WO2019023760A9

WO2019023760A9 - A control method, system and device

Info

Publication number: WO2019023760A9
Application number: PCT/AU2018/050814
Authority: WO
Inventors: Andrew King
Original assignee: Vaboss Pty Ltd
Current assignee: Vaboss Pty Ltd
Priority date: 2017-08-03
Filing date: 2018-08-03
Publication date: 2019-03-28
Anticipated expiration: 2020-02-03
Also published as: WO2019023760A1; AU2018309565A1

Abstract

A method for controlling an electrical energy consuming device comprises determining a plurality of profiles for a plurality of types of space within which the electrical energy consuming device is able to be used, where each profile defines a set of control attributes for a corresponding type of space; uploading the plurality of profiles to a controller, wherein the controller operates according to a selected one of the uploaded profiles and input of a sensor; selecting one of the profiles; receiving an input from the sensor; processing the input from the sensor according to the selected profile using the controller to determine an output signal for controlling the electrical energy consuming device; and outputting a signal from the controller to the electrical energy consuming device.

Description

A Control Method, System and Device

Field of the Invention

[0001] The present invention relates to a method, system and device for controlling an electrical energy consuming device within an environment.

Background

[0002] Cost control of electricity consumption is becoming increasing important. When an environment is not fully occupied, it may not need to be fully illuminated, heated or cooled. Zoning of heating and air conditioning is known. However, zoning of lighting is generally done on a room by room basis, and on an on/off basis. A greater degree of control of electricity consuming devices is desirable, but current implementations of this are not particularly efficient.

[0003] Several patents already exist for light control devices, such as those listed below.

[0004] US patent number 5,220,250, Szuba, now expired, uses the known method of activating lights based on detecting the presence of a person using a sensor. The sensor can be any one of various types of sensors, such as a microwave sensor, a passive infrared sensor, video or ultrasonic sensor. Each type of sensor is activated by the movement of a person within a particular area. The sensor is attached to individual light fixtures (luminaires) or is separately placed to detect movement and activate individual light fixtures. The sensor can also be placed to detect movement and activate a plurality of light fixtures which can represent zones of illumination, such as a section of a warehouse.

[0005] US patent number 7,623,042, Huizenga attempts to optimise lighting in buildings using computerised building management systems.

[0006] US patent number 9,049,756, Klusmann et al., considers lighting control in a local area network where the power is also passed from the network hub.

[0007] Published pending US patent application number 14/294,081 , Chemel et al. , addresses the management of controlled lighting through a data network and database.

[0008] US patent number 8,610,377, Chemel et al., envisages the management of controlled lighting by predictive performance through a data network and database. [0009] US patent number 8,610,376, Chemel et al., envisages the management of controlled lighting by time of day using a real time clock in order to build up historical patterns of sensor behaviour and additional services provided to and from a data network.

[0010] US patent number 8,593, 135, Chemel et al., outlines the management of controlled measurement of power consumption, with reporting and additional services being provided to and from a data network.

[001 1] Published pending US patent application number 15/094,559 Chemel, expands on US patent number 8,593, 135, Chemel et al., and adds the ability through cloud interconnection to enhance data collection, analysis and lighting control.

[0012] US patent number 8,536,802, Chemel et al., expands on the claims of several prior patents by adding in a module for data interfacing to management.

[0013] US patent number 8, 531 , 134, Chemel et al., expands on the claims of several other related patents, including US 8,536,802.

[0014] US patent number 8,457,793, Golding et al., envisages an integrated lighting and building management environment in order to infer the spatial location of fixtures.

[0015] Australian Patent AU 201 1245060 allows for inter-controller signalling, but does not address the large effort in tailoring sets of nearby lamp controllers, or other environmental modifiers, to coexist, even if they have requirements to operate in close proximity.

[0016] Whilst there are many types of energy control apparatus on the market, the more

sophisticated devices have the disadvantage that, by controlling many aspects of energy

consumption, they also require a significant amount of site analysis and control attribute design and configuration. Often these configurations values become unique to particular devices, thereby adding extra complexity to lifetime support processes.

[0017] An installer of, for example, lighting equipment in a particular area is not able to install lights in an efficient, coordinated and energy saving manner using the inventions of the prior art.

Accordingly, the present invention seeks to provide a more efficient energy consuming solution. [0018] In this specification the terms "having", "comprising" or "comprises" are used inclusively and not exclusively or exhaustively.

[0019] A reference to prior art documents is not intended to be an admission that such a prior art document forms part of the common general knowledge of a person skilled in the art of the invention in any jurisdiction.

Summary of the Present Invention

[0020] According to the present invention there is a method for controlling an electrical energy consuming device, comprising the steps of:

determining a plurality of profiles for a plurality of types of space within which the electrical energy consuming device is able to be used, where each profile defines a set of control attributes for a corresponding type of space;

uploading the plurality of profiles to a controller, wherein the controller operates according to a selected one of the uploaded profiles and input of a sensor;

selecting one of the profiles;

receiving an input from the sensor;

processing the input from the sensor according to the selected profile using the controller to determine an output signal for controlling the electrical energy consuming device;

outputting a signal from the controller to the electrical energy consuming device.

[0021] In an embodiment, the signal output from the controller is for controlling light output from the electrical energy consuming device.

[0022] In an embodiment, the signal output from the controller is for controlling heat output from the electrical energy consuming device.

[0023] In an embodiment, the method further comprises preconfiguring the controller with the plurality of profiles.

[0024] In an embodiment, the selection of one of the profiles comprises preselection of one of the profiles as a default. [0025] In an embodiment, the selection of one of the profiles comprises receiving a command from an external device.

[0026] In an embodiment, the selection of one of the profiles comprises receiving a command from a user.

[0027] In an embodiment, the sensor is at least one of: a thermal sensor, an electromagnetic sensor, a mechanical sensor, a chemical sensor, an optical radiation sensor, an ionizing radiation sensor, a biological sensor, a geodetic sensor, a light sensor, a force sensor, a humidity sensor, an air quality sensor, a camera, a payment sensor, a microphone, a Radio Frequency Identification (RFID) reader, a Hall Effect sensor, and a limit-switch.

[0028] In an embodiment, the method further comprises creation of an electrical energy consuming device layout plot to classify the area into different types of space where each type of space has a different usage for the electrical energy consuming device(s) and the associated controller(s) within the space.

[0029] In an embodiment, the layout plot provides an electrical energy consuming device installer with details of the type of electrical energy consuming device to install; where to install the determined type of electrical energy consuming device; the type of profile to select for the electrical energy consuming device and an identifier of its associated controller; and a map to record details of each electrical energy consuming device and the identifier of its associated controller.

[0030] In an embodiment, the profile is selected according to the type of electrical energy consuming device being controlled, the type of space in which the electrical energy consuming device is to be used and the type of usage of the space. The profile may also be selected according to environmental conditions of the space.

[0031] In an embodiment, the identifier comprises a serial number of the electrical energy consuming device and a serial number of the associated controller, and a unique, remote access network identifier of each associated controller.

[0032] In an embodiment, the remote access network identifier includes the media access control (MAC) address. [0033] In an embodiment, each profile comprises a plurality of control attributes, where each control attribute is suitable for configuring an element of a control paradigm for the electrical energy consuming device.

[0034] In an embodiment, selection of the profile comprises monitoring the sensor input over time and changing the selection of the profile according to patterns recognised in the monitored sensor input.

[0035] In an embodiment, the method further comprises adapting the default profile based on a recording of inputs received from the sensor over time by use of a learning algorithm.

[0036] In an embodiment, the method further comprises autonomous adaptation of the controller's behaviour via the default usage class configuration of the controller including observational and learning algorithms in the event of an electrical energy consuming device being present in a space that does not have a predefined usage class associated with the space for the controller.

[0037] In an embodiment, the controller can create a new profile from an already selected profile, by adapting the selected profile using machine learning, so as to create a new profile that matches the new environmental situation.

[0038] In an embodiment, the controller comprises software for controlling a processor of the controller that is able to perform algorithms, so that in the event that a controller is not able to identify a profile that matches any profiles present in the current database of the controller, then a new profile is able to be determined.

[0039] In an embodiment, the new profile is determined by analysing stored sensor inputs and fitting a model to these inputs in correlation with the stored profiles.

[0040] In an embodiment, the newly created profile is stored in a profile list table for later matching so that future environments with the same requirements can utilise the newly created profile.

[0041] In an embodiment, the selected profile includes an indication of the signal format that the controller is outputting. [0042] In an embodiment, the signal output from the controller can take the format of: a varying analogue voltage; a DALI, DMX or proprietary DSI wired digital control bus signal; on-off control signal for controlling relays to individually activate portions of a system under control; direct primary side proportional power control through leading or trailing edge dimming; a wireless signal using an industry standard protocol, such as Bluetooth™, Zigbee™, Wi-Fi; or a signal sent through a building management system gateway via a wire or wireless digital bus format.

[0043] In an embodiment, the sensor comprises a presence detection sensor.

[0044] In an embodiment, the presence detection sensor uses detection via at least one of passive infrared detection, ultrasonic detection and microwave detection.

[0045] In an embodiment, the presence detection sensor can be refined so as to increase the detection sensitivity of the sensor.

[0046] In an embodiment, the sensor comprises a light sensor wherein the light sensor comprises at least one of a human eye perceivable light response sensor and a colour camera element light sensor.

[0047] In an embodiment, the profile comprises at least one attribute which can be modified.

[0048] In an embodiment, the at least one attribute includes at least one of: maximum light value, dwell period, dimmed value, dimmed period, dim down rate, dim state transition rate, absolute minimum value of output, sensing threshold level, detection sample time, detection cycle time, filter parameters, sensing area, occupancy sensing size, signal detection algorithm selection, multi- sensor correlation, controller 3D coordinates, device comparison algorithm selection, device comparison rule 1 -n, occupancy type detection and rules selection, device behaviour on comparison selection, device onward signalling algorithm selection, light harvest baseline Lux, reflective light sensitivity, light power reduction rate, ambient light adjustment frequency, ambient calibration control, ambient sector quantity and location, ambient sector correlation, minimum ambient level, ambient type detection and rules section, lighting colour, time of day, time of week, temperature, air flow, humidity, gas, sound, weather forecasts, security access levels and access cut-off periods.

[0049] According to the present invention there is a device for controlling an electrical energy consuming device, comprising: at least one sensor for sensing an aspect of a region;

a controller for receiving a plurality of profiles and an input from the sensor;

wherein the controller operates according to a selected on of the profiles, or a self-created profile, and the input from the sensor;

wherein a signal output from the controller controls the electrical energy consuming device.

[0050] In an embodiment, the controller comprises a processor for creating the self-created profile.

[0051] According to the present invention there is a device for controlling an electrical energy consuming device, comprising:

at least one sensor for sensing an aspect of a region;

a processor for automatically generating a profile by fitting a model to inputs from the sensor;

a controller for receiving an input from the sensor;

wherein the controller operates according to the automatically generated profile so as to output a signal to control the electrical energy consuming device.

[0052] According to the present invention there is a system for controlling a plurality of electrical energy consuming devices, comprising:

a plurality of sensors, each sensor for sensing a respective region within the sensor's range when installed in an area which comprises the respective regions;

a controller associated with each of the sensors, each controller configured to receive a plurality of profiles;

wherein each controller receives an input from the associated sensor;

wherein each controller is configured to output a signal to control a respective one of the electrical energy consuming devices according to a selected one of the profiles and the received input from the associated sensor.

[0053] According to the present invention there is a system for controlling a plurality of electrical energy consuming devices, comprising:

a plurality of processors for automatically generating a profile by fitting a model to inputs from the respective plurality of sensors;

a controller associated with each of the sensors, each controller configured to receive a plurality of profiles; wherein each controller operates according to the automatically generated profile so as to output a signal to control the respective electrical energy consuming device.

[0054] According to the present invention there is a method for controlling a plurality of electrical energy consuming devices, comprising the steps of:

sensing a respective region determined by a respective sensor of a plurality of sensors, wherein the respective region is within the respective sensor's range when installed in an area which comprises the respective regions;

automatically generating a profile using a processor by fitting a model to an input from the associated respective sensor;

receiving at a controller associated with each of the sensors, a plurality of profiles;

operating the respective controller according to the automatically generated profile so as to output a signal to control the respective electrical energy consuming device.

[0055] According to an aspect of the present invention, there is provided a luminaire controller comprising:

a light detector for measuring illumination of an area illuminated by a luminaire;

a camera for capturing video images of the area illuminated by the luminaire; and

a processor configured to process the video images for detecting movement;

wherein the processor stores image areas where higher frequency of movement is detected over time and adjusts the sensitivity of detection within the video images according to the frequency of detection in the image areas within captured video images;

wherein when movement is detected the controller outputs a control signal to the luminaire according to a current reading of the light detector.

[0056] In an embodiment, the processor is adapted to de-focus on or decrease sensitivity on areas where movement is not previously detected.

[0057] In an embodiment, the processor is adapted to de-focus on or decrease sensitivity on areas where there is repetitive high frequency movement.

[0058] According to an aspect of the present invention, there is provided a method of controlling a luminaire, comprising:

measuring illumination of an area illuminated by the luminaire;

capturing video images of the area illuminated by the luminaire; and processing the video images for detecting movement;

storing image areas where a higher frequency of movement is detected over time;

adjusting sensitivity of detection within the video images according to the frequency of detection in the image areas within captured video images;

outputting a control signal to the luminaire according to a current reading of the light detector when movement is detected.

[0059] According to an aspect of the present invention, there is provided a luminaire controller comprising:

a processor for controlling an output signal for controlling the luminaire;

wherein the processor is configured to control the luminaire to briefly reduce the luminaire output and to measure the corresponding brief reduction in reflected light received by the light detector; wherein the processor is configured to calculate the required output control to the luminaire.

[0060] In an embodiment, the brief reduction in luminaire output is repeated so as to establish over time the luminaire output required to achieve a desired overall lux in the area.

[0061] In an embodiment, the desired overall lux is the maximum output of the luminaire in conditions of no ambient light.

[0062] In an embodiment, the processor is configured to make small up or down adjustments in luminaire output over a longer period, and to measure the response received by the ambient light detector so as to determine a response by other luminaires that produce light detectable by the ambient light detector.

[0063] In an embodiment, the luminaire controller comprises a visible light sensor and a near- infrared sensor and is configured to determine what ambient light is from sunlight and what ambient light is from other sources based on measurements of the visible light and near-infrared sensors.

[0064] According to an aspect of the present invention, there is provided a method of controlling a luminaire controller comprising:

measuring illumination of an area illuminated by the luminaire;

controlling an output signal for controlling the luminaire;

briefly reducing the luminaire output;

measuring the corresponding brief reduction in reflected light received; calculating the required output control to the luminaire based on the measured reduction in the reflected light.

[0065] According to an aspect of the present invention, there is provided a luminaire controller comprising:

a colour camera for receiving an image of an area illuminated by a luminaire;

a processor for controlling an output signal for controlling a luminaire;

wherein the processor reads an automatic brightness control value of the camera and determines the ambient light level based on the brightness control value;

wherein the processor derives spectral components average levels from the camera and analyses the spatial area for those areas that are particularly bright.

[0066] In an embodiment the analysis comprises comparing the level of red to green and blue in order to differentiate the relative sunlight to luminaire ratio.

[0067] In an embodiment, the processor is configured to momentary change the colour output of the luminaire and to process the measured change in overall spectral levels in order to determine the appropriate luminaire output by compensating for contrast between bright and dark in the brightness control value.

[0068] According to an aspect of the present invention, there is provided a method of controlling a luminaire comprising:

receiving an image of an area illuminated by the luminaire with a camera;

controlling an output signal for controlling the luminaire;

reading an automatic brightness control value of the camera and determining the ambient light level based on the brightness control value;

deriving spectral components average levels from the camera and analysing the spatial area for those areas that are particularly bright;

adjusting the output signal according to the determined ambient light and bright areas.

[0069] According to an aspect of the present invention, there is provided a luminaire controller comprising:

an image sensor for receiving an image of an area illuminated by a luminaire;

a processor for processing images of the image sensor to determine an output signal to the luminaire; wherein the processor is configured to:

evaluate a perceived effect of light output from the luminaire from the received image;

determine one or more perceived effects of light output from other luminaires from the received image;

determine a speed at which a change is made to the output and a magnitude at which the change is made based on the perceived effect of light output from the luminaire and the determined perceived effect of other luminaires;

adjust the output signal according to determined speed and effect of the other luminaires.

[0070] In an embodiment, the processor is further configured to determined responses of other luminaires to the changes to the output.

[0071 ] In an embodiment, the processor is further configured to determine a period of time over which the determinations are made.

[0072] In an embodiment, the processor is further configured to determine a number of samples to take in order to make the determinations in the determined period of time.

[0073] According to an aspect of the present invention, there is provided a method of controlling a luminaire comprising:

receiving an image of an area illuminated by the luminaire;

processing images of the image sensor to determine an output signal to the luminaire;

evaluating a perceived effect of light output from the luminaire from the received image;

determining one or more perceived effects of light output from other luminaires from the received image;

determining a speed at which a change is made to the output and a magnitude at which the change is made based on the perceived effect of light output from the luminaire and the determined perceived effect of other luminaires;

adjusting the output signal according to determined speed and effect of the other luminaires.

[0074] According to an aspect of the present invention, there is provided a luminaire controller comprising:

an image sensor for receiving images of an area illuminated by a luminaire;

a processor for processing images of the image sensor to determine an output signal to the luminaire; wherein the processor is configured to use a statistical model to:

decide the how the image area is divided into analysis sub-area based on common levels of image activity;

determine how frequently the sub-areas are analysed for movement detection based on historical activity levels in the sub-areas;

remove zero, one or more of the sub-areas from movement detection based on repetition and frequency counts that do not correlate well with the behaviour of targeted movement entities; determine an action to undertake in response to a particular pattern of movement being detected; determine an adjustment to the statistical model, if any.

[0075] According to an aspect of the present invention, there is provided a method of controlling a luminaire comprising:

receiving images of an area illuminated by the luminaire;

using use a statistical model to:

[0076] According to an aspect of the present invention, there is provided a luminaire controller comprising:

a processor for controlling an output signal for controlling the luminaire;

wherein the processor is configured to null out the contribution of natural light and other lighting devices;

wherein the processor is configured to compute the required output signal based on the nulled out contribution of natural light and other lighting devices.

[0077] In an embodiment, the output signal is computed without requiring an input specification. [0078] According to an aspect of the present invention, there is provided a method of controlling a luminaire comprising:

measuring illumination of an area illuminated by the luminaire;

outputting a signal for controlling the luminaire;

nulling out a contribution of natural light and other lighting devices;

computing the output signal based on the nulled out contribution of natural light and other lighting devices.

[0079] According to an aspect of the present invention, there is provided a luminaire controller comprising:

an image sensor for receiving images of an area illuminated by a luminaire, wherein the sensor is configured to receive the images in a plurality of spectrum channels;

a processor for processing images of the image sensor to determine an output signal to the luminaire;

wherein the processor is configured to determine a contribution to illumination in the received images by the luminaire by changing the output signal;

wherein the processor is configured to evaluate the changes in output of the spectrum channels of the image sensor;

wherein the processor is configured to evaluate non-luminaire illumination contributions by evaluating the difference in a measured response to that of the measured multi-spectrum response to the changes in output to the luminaire.

[0080] According to an aspect of the present invention, there is provided a method of controlling a luminaire comprising:

receiving images of an area illuminated by a luminaire, wherein the images comprise a plurality of spectrum channels;

determining a contribution to illumination in the received images by the luminaire by changing the output signal;

evaluating the changes in output of the spectrum channels of the image sensor;

evaluating non-luminaire illumination contributions by determining a difference in a measured response to that of the measured multi-spectrum response to the changes in output to the luminaire.

Description of the Drawings [0081] In order to provide a better understanding of the present invention, preferred embodiments will now be described, by way of example only, with reference to the accompanying drawings in which:

[0082] Figure 1 is a schematic view of a control device of the present invention;

[0083] Figure 2 is a schematic view of a control device within a control system of the present invention;

[0084] Figure 3 is a diagrammatic representation of an undesirable ambient light feedback scenario occurring;

[0085] Figure 4 is a flow chart illustrating a process followed when a controller has default sensor profiles uploaded that allow a particular sensor profile to be selected;

[0086] Figure 5 is an example of a location plot layout;

[0087] Figure 6 is an illustration of two tables which indicate the number of the plot from Figure 5, the associated power of the LED chosen for the plot from Figure 5, and the type of activity assigned to the plot from Figure 5, where the types of activities are defined in the second smaller table;

[0088] Figure 7 is an example of vehicle presence detection by a sensor and the associated digital signal processing analysis that occurs in an embodiment of the invention;

[0089] Figure 8 is a block diagram of adaptive learning occurring in an embodiment of the control system of the present invention;

[0090] Figure 9 is an example flow chart of a luminaire (light) control system illustrating adaptive learning of the control system through detecting presence via a sensor and learning from the presence detection;

[0091] Figure 10 is an example flow chart of a luminaire (light) control system illustrating adaptive learning of the control system through ambient learning via, for example, ambient light; [0092] Figure 1 1 is a diagram illustrating how detection events are correlated and groups are formed;

[0093] Figure 12 is an example flow chart of a luminaire (light) control system illustrating adaptive learning of the control system through group control learning via, for example, communication between controllers;

[0094] Figure 13 is a flow chart illustrating an example of a control system according to the present invention.

Description of Embodiments of the Invention

[0095] Figure 1 shows a control device 10 comprising a controller 12 and a sensor 14. The sensor 14 is arranged to detect features of a sensing region 18. For example, the sensor 14 may detect a person within the sensing region 18. The detected features are input to the controller 12. The controller 12 produces a control signal 16 to an electrical energy consuming device 20, such as a light emitting device, so as to control an output 22 of the electrical energy consuming device 20, which, as shown can be light. The control signal 16 is determined by the controller 12 according to a selected profile of attributes that determine the manner of control of the electrical energy consuming device 16 according to the detected features received from the sensor 14. The attributes are parameters that the controller 12 uses to determine how to control the energy consuming device 20.

[0096] Figure 2 shows an embodiment of a system 100 comprising a plurality of devices 10A and 10B, a plurality of energy consuming devices 20A and 20B, and each device having a sensor 14A and 14B, respectively, sensing a corresponding sensing region 38 and 40.

[0097] In such a system 100 the output of the respective energy consuming devices 20A and 20B can be subjected to many differing conditions. Just one of these is feedback. In this figure there are two possible types of feedback occurring. The first possible feedback situation occurs when the first sensing region 38 overlaps with the second sensing region 40. The overlap between the sensing regions can result in 'false' sensing occurring due to the overlap and not when an actual detection event occurs. This can result in the outputs of the first and second electrical energy consuming devices 20 and 20B changing in accordance with the 'false' sensing resulting in unwanted feedback occurring. [0098] The second possible feedback situation occurs when the first sensing region 38 does not overlap with the second sensing region 40. Whilst the sensing regions are not overlapping, the range of the outputs of the first and second electrical energy consuming devices 20A and 20B are overlapping, and this again results in 'false' sensing detection occurring.

[0099] These feedback situations are shown in the diagram of Figure 3.

[0100] In order to accommodate the variety of environmental conditions, including, but not limited to preventing such feedback from occurring, one embodiment of the present invention aims to accommodate the different environmental conditions in the set of profiles loaded onto a memory device of the controller 12. Each profile configures the controller 12 to operate differently so that the electrical energy consuming device 20 can be efficiently used according to the circumstances of the environment of the device 20 at a given point in time.

[0101] In another embodiment, the controller 12 adaptively learns from inputs of the sensor 14, so that if there are no profiles that match the requirements for a space, then a new profile can be automatically generated through fitting a model to the sensor 14 inputs and adaptively learning and generating a new profile accordingly. This embodiment is described in more detail further below.

[0102] In an embodiment of the present invention, a memory of the controller 12 within the device 10 is uploaded with a predetermined set of profiles (not shown in Figure 1 ) for each type of space defined within an area, where a type of space could be, for example, a passage way located within for example, a warehouse, where the warehouse is the area. After the controller 12 is uploaded with the predetermined set of profiles, a profile is selected. Each profile has control attributes defined within it, which are specific to the particular profile.

[0103] Once a profile is selected, the control attributes within the profile, along with the information received at the controller 12 from the sensor 14, are used to control the output 20 of the electrical energy consuming device 20, which, as shown in the example of Figure 1 , can be a light emitting electrical energy consuming device.

[0104] The sensor 14 is used to detect, for example, a person, within the sensing region 18. The detection is then relayed to the controller 12, so that the controller 12 can then apply the control attributes within the selected profile in the control signal 16 to the electrical energy consuming device 20 to emit, in this example, light 22, in accordance with the information received from the sensor 14. [0105] Figure 4 illustrates a process for configuring a controller 12. The first step is to assess physical aspects of the site. There are many aspects of a location that can influence control attributes; there is for example the physical site structure, presence of skylights, orientation of such skylights, light reflectivity of surfaces, insulation of surfaces, or height of ceilings.

[0106] Next, the usage of the site is assessed. The site usage aspects have large influences on environmental control such as; occupancy and concentration of workers in areas, type of work and hence needed illumination levels, allowable temperature ranges, and ventilation needs.

[0107] Next, any special needs of the site are assessed. These can range from; clean room or sealed areas, dust free or dust generating spaces, dangerous environment controls, external environmental influences such as contrast minimization when direct sunlight hits localized spots and fail-safe requirements.

[0108] User preferences are assessed. User preferences could range from time-of-day changes to control attributes -e.g. a skeleton second shift needs a different temperature / lighting profile;

whether controlled lights turn fully off or remain at low levels under high ambient or low occupancy situations; what level of control adjustment is able to be done by on-site staff; and the level of integration of control with third party systems.

[0109] Professional knowledge about site requirements is assessed. Professional knowledge could range from the specific location of environmental systems and their control: lighting, HVAC, sound control to process design layout and human factor design for productivity and wellbeing. These design influences can have a direct influence on control locations and profile settings.

[01 10] These assessments are then used to generate multiple configurations of control profiles for the controller 12 to use according to sensory input detected by the sensor 14, as indicated in by the top table in Figure 4. The controller 12 is pre-loaded with a full set of default profiles. Of the preloaded profiles operating control profile(s) are selected, as indicated by the middle table in Figure 4.

[01 1 1] In one embodiment additional use case data is added the selection model. In an embodiment, in certain circumstances, an individual attribute is adjusted to account for these circumstances, as indicated by the bottom table in Figure 4. [01 12] The specific behaviour of the controller 12 in relation to the device 40 is based on the settings of a wide range of control attributes relevant to the sensor types connected to it. The preprogramming of profiles would without the present invention be a tedious and error-prone task of manual programming. Manual programming would also place considerable cost and effort onto the supplier for training and post install support. In many cases this additional complexity has limited the energy savings and environmental fit of controllers, because of the incentive to minimise the configuration interface complexity.

[01 13] The settings provided in the profiles are arranged in such a way that the best controller 12 behaviour for a particular location and environmental use can be selected by sending to the controller 12 one value for each behaviour type. Alternatively, the controller 12 can be equipped with the additional sensing and learning algorithms that, in effect, allow it to self-determine the best behavioural setting, as described further below.

[01 14] The invention solves this error-prone and onerous process of manually configuring sophisticated energy controllers 12 by embedding in the controller 12 a wide range of highly tailored performance profiles. These profiles can be selected in advance through detailed usage model templates or alternatively, by providing the controller 12 access to additional sensor capability and machine learning algorithms. It also allows, after a best fit profile is selected, to fine tune individual attributes or even a change to a new, multi-attribute profile.

[01 15] The controllers 12 are able to manage environmental control outputs via controllable drivers from a wide range of suppliers and protocols. The required mechanical and electrical interfaces are provided via a supporting electronic interconnect module that provides ancillary services such as complete on-off switching and power conditioning to controllable devices that are not able to provide it themselves.

[01 16] By creating a comprehensive model of a number of recurring occupancy usage patterns, together with detailed analysis of all of the control attribute values most suited to that usage, it is then possible to pigeonhole most environments needing energy control into one of a small number of standard "profiles" for each control type: e.g. occupancy, temperature control, light level.

[01 17] In addition, even if these configuration processes are carefully designed and documented, they cannot deal with many instances of changing occupational needs such as:

• New building installations where the end-use user or activity has not been determined; • Refurbishment of premises between vacancies where the next tenant has not been determined;

• Project based areas in warehouses, factories and office environments, where the occupancy usage may change over time;

• In general purpose activity centres where usage may change throughout a day;

• Installation by organisations that have indirect and unknown knowledge of device

configuration.

[01 18] Site-specific factors that contribute an additional degree of difficulty are:

• Roller doors that are auto opened to excessive height;

• Roller doors that are manually opened to varying heights by staff working variable shifts;

• Roller doors with climate friendly/unfriendly orientations throughout four seasons of the year;

• Moving pools of direct sunlight where the average illumination level needs to increase

because of the need to reduce contrast between sunlit and other areas;

• Skylights with climate friendly/unfriendly orientations throughout four seasons of the year: o Both North and South pitched roofs receive maximum permissible ambient light in the southern hemisphere summer;

o North pitched roofs receive slightly reduced ambient light in the southern hemisphere winter;

o South pitched roofs receive considerably reduced ambient light in the southern

hemisphere winter;

o East and West pitched roofs receive ambient advantages in the morning and

afternoon respectively,

• Open plan large format assembly areas with constantly changing surface finish and

reflectivity such as:

o Prefabricated building;

o Vehicular manufacture;

o Aircraft repair and maintenance;

o Sporting arena;

o Hard stand open storage,

• Renovation to structure due to tenancy variation, e.g. opaque skylight renewal.

[01 19] The current most advanced lighting controllers generally require a degree of manual adjustment. Adjustment can be either: • Observation with the /an operator analysing conditions and implementing discretionary adjustment to base settings; or

• Observant - when certain benchmarks are reached incremental adjustments are prescribed and affected.

Some embodiments of the present invention do not require this.

[0120] Whilst promoting the concept of light control with configurable default settings there are certain behaviour factors that contribute to the need for an additional degree of precision energy control:

• Rotating shifts;

• Lack of incentive / interest of on-site personnel e.g. If an area manager or operator does not benefit from savings to be made from non-core initiatives, it is unlikely that data will be dutifully collect and analysed so manual adjustments can be made to realize optimum benefit to the system;

• Lack of knowledge:

o The skill required to adjust attributes for the first major portion of energy savings

should be within the capability of persons of average skill. That could be the first 80% of energy saving;

o The skill required to generate the balance or final 20% will require an understanding of effect of beam angles, air currents, insulation / reflectivity of materials, diffusers, CRI, Kelvins Colour temperature, etc. That level of skill is not generally present in the average business or community premises.

[0121] In an embodiment, the controller 12 is a system on chip (SoC) device, which contains:

• Powerful, low power, 8, 16, 32 or 64bit processors;

• At least hundreds of kilobytes of programmable persistent memory;

• At least tens of kilobytes of volatile random-access memory;

• Diverse and powerful peripheral functions such as digital-to-analogue converters, real-time clocks, flexible timers and digital I/O ports.

[0122] Some SoC devices also come complete with high performance, short-range, industry standard wireless transceivers and protocol stacks. [0123] Figure 5 illustrates a site plan of a building in which there are two types of Luminaire. The first type is a 150W LED - indicated in red. The second type is a 100W LED - indicated in blue. The layout of the each type of luminaire is shown and is labelled, in this case with a number.

[0124] Figure 6 illustrates how the location plot of Figure 5 can be used to classify spaces into different usage types associated with each electrical energy consuming device and their respective controller. The naming and determination of the various usage types is determined prior to the classification of spaces occurring. Generally, the naming and determination of the various usage types occurs by, for example, discussion and determination between a lighting designer and the building user, in this example.

[0125] The naming and determination of usage types for the spaces, provide the installer of the electrical energy consuming devices with the coordination of: the type of electrical energy consuming device to install and where to install it; the most suitable profile to select within the controller, for the location of the electrical energy consuming device; and a map to record the serial number(s) of each electrical energy consuming device and its associated controller as well as the unique remote access network identifier (e.g. MAC address) for each controller.

[0126] The selected profile can be determined from a specific usage type (e.g. Pallet Rack Aisle; General Work Area; Inwards Goods) which has been previously determined by the lighting designer and building user.

[0127] After a usage type has been assigned to a space, the electrical energy consuming device and its associated controller that are located within the space, will be controlled by the profile having the control attributes associated with the usage type assigned to that space (i.e. the controller will be controlled by the profile selected within the set of profiles uploaded to the controller).

[0128] The uploading of profiles to the controller can be completed by pre-configuration, so that a specific profile can then be selected in a number of ways. The first way in which a profile can be selected is via the relevant profile being the default profile when a stored program is loaded into the controller. Another way is through wired communication to the controller, once programmed (for example by an UART command sequence or a proprietary extended DALI bus command). Another further way is by wireless command (using a proprietary Wi-Fi command sequence or a change to a custom Bluetooth service UUID). [0129] Each profile has control attributes associated with it, and these attributes are developed using a specific methodology in order to: maintain conformity with any government or workplace standard; provide repeatable results when similar occupancy areas are analysed; allow a continuous improvement of control attribute optimisation as a result of post implementation reviews of novel location usage types.

[0130] The development of a profile in this embodiment, is via a process which involves incorporation of minimum occupational safety and health (OSH) standards for work or occupancy types; formalising a questionnaire as a result of surveying building users across a wide range of activities; consulting with professional energy management entities that have had extensive experience in the specification of effective energy management in corporations and public facilities; receiving input from lighting level modelling packages; developing issue specific user surveys.

[0131] Each of these components is then used as inputs to develop a profile. The inputs are quantified and an interactive analysis spreadsheet designed. The spreadsheet is then used to generate a profile by correlating and fitting a model to the data across the diverse range of site and usage types.

[0132] As a result of sensory inputs, attribute settings and decision-making rules within the controller 12, it is able to issue energy control commands in a wide range of formats. These include:

• A varying output voltage, commonly called 0-10V or 1 -10V analogue voltage control;

• The DALI, DMX or proprietary DSI wired digital control bus formats, used to send commands to a set of one or more devices within a controller device's domain;

• A simple on-off control to a set of one or more relays on a primary power circuit that

individually activate portions of the device under control;

• Direct primary side proportional power control through leading or trailing edge dimming;

• A wireless communication interface using one of the industry-standard protocol stacks such as Bluetooth™, Zigbee™ or Wi-Fi;

• To indirectly control devices through a building management system gateway via one of the wired or wireless digital bus formats.

[0133] The sensor 14 may be one or more of:

• a thermal sensor;

• an electromagnetic sensor;

• a mechanical sensor; • a chemical sensor;

• an optical radiation sensor;

• an ionizing radiation sensor;

• a biological sensor;

• a geodetic sensor;

• a light sensor;

• a force sensor;

• a humidity sensor;

• an air quality sensor;

• a camera;

• a payment sensor;

• a microphone;

• an RFID reader;

• a Hall-effect sensor; and

• a limit-switch.

[0134] This controller 12 will require one or more characteristic values to be defined in order to control the device 20. The control characteristics could have attributes, such as:

• The temperature at which a controller 12 makes an output change;

• The light level at which a controller 12 makes an output change;

• Time of day/week/year, that would affect the controller's 12 decision to make an output change;

• Elapsed time of an unoccupied space that would cause a controller 12 to make an output change.

[0135] This list of control attributes can become very extensive (for example, precision control of a single Ambient Light and Occupancy Controller can easily require more than 20 control attributes to be set). In this case the training load to enable luminaire installers or users of controlled spaces (people not skilled in the art of precision control systems) to properly understand each control attribute adjustment in controller not capable of performing as the present invention does cannot be justified.

[0136] A simpler and more accurate method of configuring all of the control attributes of such the controller 12, so that the energy usage is minimised whilst the user benefit of the energy consumption is not adversely impacted is provided by the present invention. The present invention achieves this aim by defining a series of control attribute settings as an occupancy usage type "Profile".

[0137] Each "Profile" is created by in-depth analysis, and selection of control attributes that best match of the user requirements of a particular space. This space could range from, for example:-

• Aircraft hanger;

• School Corridor;

• Community Hall;

• Pallet rack aisle in a warehouse;

• External or underground car park.

[0138] Now, instead of an installer or equipment specifier needing to determine each attribute, all that is required is for them to select the best "profile" that matches the space usage.

[0139] For sites where there is much less direct knowledge of the optimum usage type, or where the environment is subject to change, this invention defines how a default usage configuration that can be equipped with observational and learning algorithms to allow the control behaviour to evolve over time in an autonomous fashion without reliance on network connectivity.

[0140] The first learning method requires that a wide range of activity areas are analysed as to define the types and values of energy control "profiles" that are a best fit for their needs. This core model is used to configure all the control attributes of the environmental controller, by selecting a single profile in each controller out of an arbitrarily large configuration data set that has been preloaded. For example, a controller that regulates light output in response to occupancy and ambient light sensing would require just two, single number, profile selections- one in the set of possible occupancy types and one in the set of ambient control responses.

[0141] In the second learning method, the controller 12 is equipped with additional environmental sensing and learning algorithms so as to derive a best fit "profile" itself, this then allows the controller to select the optimum set of control attributes.

[0142] In this embodiment of the invention, the optimum or "best fit" profile selection does not prevent further fine tuning of individual control attributes. The controller is able to operate completely autonomously with either a 1 :1 , or 1 :many relationship between the controller 12 and its controllable energy consuming devices 20. [0143] Once a profile has been selected, either by default settings, or by a wired or wireless connection to the controller 12, its autonomous operation does not preclude the ability to pass information through network services, but it is not a prerequisite for operation of the system architecture.

[0144] In addition, this autonomy is enhanced by the controller's ability to operate from a small amount of power directly from a source in the controlled device, e.g. a LED array or other DC supply. This very low power sourcing is very flexible and can range from 5 to over 150 volts.

[0145] Figure 7 provides an example of adaptive leaning control. The controller 12 receives input from the sensor 14, which is a standard presence detection sensor that uses, for example, passive infrared, ultrasonic or microwave detection, can be refined in regard to increasing its sensitivity, by analysing signals at an analogue level before they are passed to a threshold detection system.

[0146] Standard sensor detection algorithms for these types of analogue sensor typically provide a simple uni-directional or bi-directional signal threshold detection level; giving a simple yes / no control signal to the controller.

[0147] According to an embodiment of the present invention, a more advanced detection algorithm can apply digital signal processing to the raw sensor signal to extract further information. In the case of sensor outputs that comprise both amplitude and frequency information, a Fourier analysis-type of algorithm may be able to provide some information as to both the apparent size and relative speed of a detected object. This estimated size information may be used to determine whether this detection event should be acted on.

[0148] In order to achieve refinement of control, in a general case of analogue-type sensors, the output of the analogue sensor 14 is continuously measured and stored so as to be able to detect when a detection threshold is detected. Once a detection threshold is detected, the stored 1 to 2 second signal pattern of the sensor is analysed by a correlation function against all of the similar previously stored sensor signal patterns. Once the correlation value reaches a certain threshold, the controller then acts on the presence that has been detected. The controller therefore takes action before a full threshold value is detected, which effectively increases the sensitivity and response time of the detection process of the sensor. It may not be possible to implement this capability in a stand-alone sensor, as a stand-alone sensor may be subject to local environmental noise, as well as presence detection signals. [0149] A further embodiment of the present invention involves the controller having the capability to store detection event patterns in time sequences. For example, there could be: an isolated single detection event separated by long periods of no activity; alternatively, there could be a strong cluster of presence detection events followed by a weakening frequency; or the opposite pattern of slowly increasing detection frequency; or a continuous pattern of medium level detection events. Each of these pattern types can be assigned a signature profile or usage type. In the examples cited, the controller 12, given that it is configured with a warehouse location type, it can then infer that the first sparse pattern indicates location in a pallet rack aisle; the second pattern suggests location in an inwards goods area; whilst the third pattern suggests location in an outwards goods area and the last pattern infers a general work / administration area. Accordingly, the optimum dwell, dim level, dim to off, dim down and dim up periods could be set / reset to these selected profiles that it has stored in its reference library.

[0150] A further embodiment of the present invention employs a low cost, high resolution, VGA camera module, (or similar) as well as the primary ambient light detector as the sensor 14. The relative high resolution and potentially long range of vision systems offer potential advantages over other detection sensor technology such as passive infrared or microwave. These devices are commonly used in security cameras, and, in some instances, use software algorithms to detect movement in the vision field which can then issue alarms or recording activation. A common limitation of these devices is how to exclude natural movements (such as wind-induced) from creating false positive events. This effect can be minimised by manually altering the field of view through a visual user interface for every camera image. This heavy customisation load is not practical for low cost. In the present invention, adaptive learning energy controllers with an alternate detection methodology can be employed. In such instances, the controller will use a range of edge detection and colour filter algorithms to achieve an initial capability of identifying movement within its field of vision, but, over time, its ability to detect movement will increase. The increased sensitivity will arise due to its ability to store the general coordinates of where each presence detection event occurred, and the controller 12 will be able to determine all of the field of view areas where presence was never detected and therefore focus on the areas where activity has been detected. Alternatively, the controller will also be storing area coordinates where repetitive, high frequency movement is detected and removing focus from these areas as well.

[0151] A further enhancement on the previous embodiment of the present invention is for the controller 12 to keep two different optimised fields of view. One field of view for before a detection event and the other field of view for when it has activated the device 20 and is looking for evidence of continued occupancy in order to keep the activation timers running.

[0152] The controller 12 could be configured so that in the unoccupied state, sensitivity is increased in the peripheral areas of view, whilst during the occupied state, increased sensitivity occurs in the more central areas of the field of view.

[0153] There is also the possibility for a processor within the controller 12 to generate a new profile. Figure 8 illustrates the automatic generation of a specific profile for a space, where, for any of a range of reasons, the most appropriate profile selection cannot be made. Some possible reasons for not being able to make an appropriate profile selection are that a controller has not yet been installed, or that the usage patterns of a space may vary too frequently to justify periodic adjustments of each controller within the space.

[0154] In order to accommodate for these situations, each controller can be equipped with software algorithms which control the processor to compute, generate and select a new profile, through analysis of all current and stored sensor input data. The new profile is then input into some of the controller's attribute settings.

[0155] In order to arrive at the new profile, the flow that is undertaken is illustrated in Figure 8. As shown, Step 1 of Figure 8 has the controller being initialized with default attributes. After initialisation, the controller establishes a real-time servicing (Step 2) of its configured sensors (the arrangement of which the controller is able to self-determine) (Step 3).

[0156] Some sensors, such as raw microwave, passive infrared and video, require high speed, specialised signal processing algorithms. The algorithms for these types of sensors can be carried out on the controller's main processor or performed by a dedicated co-processor. For example, an ambient light sensor subsystem (3a) provides sensor information to a presence detection digital signal processor subsystem (3b) and/or a video system ambient and presence detection subsystem (3c) is used to detect a presence. When a sensor detects a presence, termed a sensor detection event, this event is passed from the sensor subsystem (Step 3) to the learning subsystems (Step 4), and the event can also be passed directly to the appropriate event-driven subsystems if an immediate response is possible. [0157] The learning subsystems (Step 4) will then take the sensor event, timestamp it (Step 4a) and enter it into a correlation analyser (Step 4b) which will then compare it with later or preceding events, in order to establish a new pattern profile (Step 4c). This new profile is then stored in a statistical analysis framework (Step 4d) which over time assigns probabilities to the event reoccurring at certain frequencies in a persistent pattern store (Step 4e) of learned event trigger references.

[0158] Once the reoccurrence probability has reached a threshold, the pattern profile is passed to the associated events table (Step 5), where every new sensor event sequence is loaded into. The pattern profile matching may also, or alternatively, be sent to the control attribute manager (Step 7) where, depending on the particular learning rule, the pattern profile match may cause changes to control attributes in both the sensor subsystems (Step 6a) and in various event-driven subsystems (Step 6b). Any change in control attributes during the matching process, is recorded in the reporting subsystem (Step 8), so that the current configuration of the controller may be available for recovery or replacement purposes.

[0159] At some stage of a sensor event pattern matching process, the attribute in the relevant learned association table row (Step 5), indicates that an event has been detected or will be detected (based on the probability profile of that learned event). This event is then treated as a normal event and signalled as such to the real-time scheduler (at Step 2).

[0160] As shown at Step C in Figure 8, non-core sensor options, the adaptive learning support is also able to be used for additional sensor types such as: temperature sensors, sound sensors, trace gases sensors, humidity sensors.

[0161] These additional sensors can either contribute to the learning algorithms of other sensors or be configured to control other environmental activators such as blinds, dampers, split system air conditioners, and vents. These activators can either be directly controlled by their associated controller or remotely controlled via a communications interface to network connected elements.

[0162] The benefits of integrating these additional sensory inputs into a common learning framework are that a simpler overall environmental management system is created. Additionally, there is the capability of being able to have an increased accuracy in the profiling of energy use, through better modelling of environmental needs, and broader sensory awareness. [0163] As shown in Figure 9, the adaptive learning of Figure 8, can be readily applied to an example that uses an electrical energy consuming device of the form of a light emitting device (luminaire). The processor of the controller is used to generate the new profiles via the process outlined in Figure 8, with the additional usage of a long term distribution table and a daily distribution table. In a minimum implementation form of this invention, the controller may not have continuous (24x7) operation and may not possess a real-time clock - with all of the system complexity and costs associated in ensuring that such a clock remains in synchronism to external time. If, in a typical scenario where power may be applied for an operational period each day, it is undesirable for the learning algorithm to start from scratch each power cycle (e.g day).

[0164] The long-term table values are designed to slowly evolve over time to take into account such influences as seasonal changes and evolving work practices and space utilization. The controller is able to keep count of how many hours it is "on" for in each cycle and just before the completion of this time period it modifies the long term table values, by a percentage of something between 5% and 20%, based on the length of its "on" period and how much variance there is between the long and short tables.

[0165] Similar to the multiple electrical energy consuming devices example of Figure 2, which has multiple controllers associated with respective multiple electrical energy consuming devices, another example of this situation can be to use a controller to control ambient light present, instead of artificial light from a light emitting electrical energy consuming device. In this example, the only role of a controller is to attempt to keep illumination across a broad area containing many controllers (where each controller controls an associated electrical energy consuming device) above a minimum lux level regardless of the amount of daylight distributed across the area.

[0166] In order to control the ambient light, it was realised that it would not be practical to have a controller follow a highly responsive real-time electrical energy consuming device (luminaire in this example) adjustment algorithm, due to the massive potential of positive feedback occurring across a large number of the controllers.

[0167] For example, if controller A in Figure 3 reduces the output of its associated electrical energy consuming device (a lamp in this example), the adjacent controller B senses its overlapping light output region has been reduced, so controller B increases its lamp output accordingly. This in turn causes controller A to further reduce its lamp output and so on. Additionally, as controller C also shares some of its light output region with both controller A and controller B, then controller C in some instances will also be applying positive feedback back to the sensing region of controller A, as shown in Figure 3.

[0168] A further complication for dynamic light control is sudden transitory light increases such as welding or flash lamps, or a temporarily parked fork or truck which has an operating rotating warning lamp, when you would otherwise not have expected the ambient illumination to change. Additionally, as each lamp typically has several neighbours contributing to a commonly lit light output region, the feedback models can become very complex. In particular, the feedback models can continuously vary in many different lit spaces as various object types with vastly differing reflectivity levels, move through the lit spaces. In order to address this problem, an embodiment of the present invention employs a set of controllable attributes which are a minimum set of attributes.

[0169] The minimum set of attributes are: Maximum light output (% of total luminaire and total Lux); Ambient light threshold (the level above which lamp output reduction occurs); Average of floor light reflectivity index (light sensors look down so that the reflectivity of the sensing area is required to compute radiant flux); Ambient sensor sample count (average of many readings is required to filter out short term events and to damp down instability); Baseline power reduction rate (how much light reduction and at what rate should a change in sunlight cause); Ambient minimum level (should high ambient readings cause complete switch off or should residual light remain).

[0170] Ambient Control only makes sense when artificial light is being used to supplement natural light in an area (in a fully enclosed space, such as cool room, there is no requirement to calculate the supplementary light requirement). Calculating the necessary supplementary light level is complex as it requires attributes such as:

• the minimum ambient level,

• surface reflectivity,

• over what time range measurements are made, typically it has been found that a sampling period of 10 - 30 seconds can minimize short term effects as camera flashes, welding, etc whilst reacting sufficiently fast to direct sunlight with rapidly moving small clouds ;

• over what time period adjustments are made, this period tends to mirror the chosen sample rate so that the system is stabilized for the start of the next sampling period;

• what rate of adjustment for each measured lux deficiency, this rate matches the adjustment period, so that the system is stabilized for the start of the next sampling period; and

• the need to take into account the influence of adjacent lamp outputs and their response. [0171] An example of the generic algorithm for this process is:

1 . Lux measurements are made by the controller once every time period (e.g once a second);

2. After a number (say 16) measurements, the average value lux is derived;

3. This value is adjusted in response to the "reflectivity" value of the site, which is built into the selected profile, so that the light level at eye level can be computed;

4. This "eye level" value is compared to the ambient lux target of the selected lighting profile;

5. A visual indication is made as to whether the Controller has calculated that there is too much or too little light in order to assist the commissioning process, if individual light level tailoring is required;

6. A calculation is made as to what the offsetting power reduction value for the Controller for each plus or minus lux requirement change. This could be a plus or minus change to the value that was previously arrived at in the previous cycle. The equation is controlled by a value defined within the chosen ambient control profile;

7. This new, calculated power offset value updates to the previous value;

8. The difference between new and old values is divided by the number of level changes that occur in each power level adjustment cycle;

9. Once this adjustment process is complete, the ambient sampling cycle begins again.

[0172] This learning embodiment of the invention, attempts to minimize the number of control attributes down to a single figure - the required overall lux level. It does this by exploiting 2 characteristics of LED lighting:

• Very fast response;

• A defined spectrum significantly different to natural light.

[0173] In order to produce a controller that can be installed without any predefined knowledge of what ambient light level it needs to maintain, it can be equipped with a leaming algorithm that initially makes only one assumption -that it has been deployed to control one element of a luminaire matrix of such a power and spatial arrangement that, on full power and with no natural light contribution, at its full output it will achieve the designed illumination level. Therefore by continually tracking the lowest measured light level, when it is operating at full power (that being the situation when there is no natural light contribution), it will able to determine the setting that should be used as the ambient control benchmark. The controller then continues, now knowing its overall Lux setting, by deliberately reducing its light briefly and measuring the received lux, it can calculate what the area "reflectivity" is. By repeatedly doing this sampling, it can null out any effects of sunlight variations and changes in other light fittings. Also by adjusting up/down, a small amount for a longer period, it can measure the response, within its field of view, to other LED lamps.

[0174] The above implementation may require some extended time period to reach the lowest value (no natural light contribution) if it is only operated in daylight hours and installed in the summer season. In a more sophisticated implementation, it is equipped with calibration tables for a visible and near-infrared sensor that allows it to estimate, at any lux level, what is from sunlight / other sources and what is from LED lighting. As it knows at any measured lux level its contribution, it can calculate the proportion of light in its viewable area that is generated elsewhere. This measurement is effectively the "gain" value of multiple inter-related lamp fixtures, and the controller will use this value to determine, how aggressively and how quickly it will react to measured light changes.

[0175] In the example where it has determined that it is isolated - i.e. there is no other LED lamp illuminating its field of view, it can respond to lux changes with a gain of "one" and a speed of "immediate". In the opposite case where it has determined that is only contributing to say 10% of the measured LED light, it will respond with say 10% of the required light change on the assumption that all other related controllers were acting in the same way. Also in this case the speed of response will be much slower as to prevent sudden reactive changes to light levels and to network feedback instability.

[0176] In a further implementation of this invention, the use of a low-cost image sensor will give an accurate, overall light level reading, but will also allow the illuminated image area to be segmented to look for areas of very high brightness that correlate to direct sunlight pools. If this situation is sensed, the controller can choose to over illuminate the area in order to minimise contrast between the sunlight pools and the artificially lit areas.

[0177] Whether a controller fully switches off a lamp under excessive ambient conditions is actually a User Preference factor in profile selection as it has been found by experience that some user groups are more comfortable in knowing that the lamp is still operational - at say 5% to 10% in this situation.

[0178] When a human eye response-type sensor is controlled by this minimum set of attributes, it allows ambient light control to operate effectively across most of the relevant spaces without instability, and the sensor is also able to deliver close to the optimum amount of additional lux to compensate for natural or other light source changes. [0179] Extending on this ambient light control example, Figure 10 illustrates another embodiment of the present invention in flow chart form, in which a controller has at least a human eye response- type sensor and an additional near infrared sensor. The controller is also equipped with

measurement and analysis algorithms that dynamically adjust any of the minimum attributes outlined above which are determined by the controller to be sub-optimum, so as to maintain energy efficiency within the system.

[0180] When the human eye response-type sensor was controlled by the minimum set of control attributes, it allowed ambient light control to operate effectively across most of the spaces without instability and was able to deliver close to the optimum amount of additional lux to compensate for natural or other light source changes.

[0181] Now, by incorporating an additional near infrared sensor, dynamic adjustment can be achieved by comparing the near infrared sensor reading with the human eye-like sensor response. As modern LEDs produce very little near infrared compared to sunlight, it is possible for the controller to calculate what proportion of its received light levels are due to itself or adjacent luminaires. This knowledge is further enhanced by momentarily dropping the luminaire output to zero (or a low value) for an imperceptible period of milliseconds and calculating the difference in sensor reading, and therefore its own contribution, to the lux level.

[0182] Once these calculations have been performed, the controller will then 'know' what the level of average incident external natural illumination is, what other electrical energy consuming devices such as luminaires are contributing, and what the controllers own associated electrical energy consuming device (luminaire in this example) contribution is.

[0183] Collision between signaling of controllers can be avoided because the sampling periods are very short, there will be natural variations in sampling periods as a result of non-rigid periodic job scheduling and each measurement is averaged over a large number of cycles.

[0184] Now that the controller has the total lumen output of the associated electrical energy consuming device (luminaire) and how the lumen output is measured, the controller can then be used to compute the most likely height / area reflectivity index. Then, from this data, the controller can then decide for each lumen of light below the threshold, the magnitude of the controller's compensatory response so as to control the output of the electrical energy consuming device accordingly. [0185] In addition, by measuring variations of incident artificial light that fall into the sensors sensing area, the controller can determine the stability of the localised network response to changing light levels, and from this measurement, the controller can dynamically adjust the size of its own response to the changing ambient light levels.

[0186] A further embodiment based on the embodiment of Figure 10, is when the controller uses a low-resolution colour camera element as its light sensor. At a macro level, the ambient light level can be read from an Automatic Gain Control register of the camera, but further use of the camera can be achieved by deriving both the red, green and blue light components (RGB) average levels and then analysing the spatial area for those areas that are particularly bright.

[0187] The analysis of R to GB relative levels can be used to differentiate the relative sunlight to LED lighting ratio. Compensation for its control of differing colour temperature lamps can be made by observing the change in response on each colour channel by a momentary change in output of the lamp and simultaneously measuring the relative change in the overall RGB values. Accordingly, the analysis of bright areas can cause the relevant controller to increase its light output level over and above what it would have computed from the AGC signal alone, because of the desirability to reduce contrast levels between bright and dark areas in the space's lit region.

[0188] An additional benefit of using a low-resolution colour camera element sensor is that if a controller is alerted that it has direct sunlight in its field of view, then even without a real-time clock, the controller would be able to learn high brightness patterns and develop a view of the time of day and the day of the week. This knowledge can then be fed to the processing streams of other sensors, if fitted.

[0189] The image sensor field of view is divided into an n-by-m matrix and the brightness level for each of the RGB channels is averaged for each matrix element. The size of the n-by-m matrix is fundamentally determined by the available processing power of the controller's processor but can be in the range of 4-by-4 to 16-by-16. The choice of matrix size is site-specific and could be for example 6-by-1 for a pallet rack aisle and 8-by-8 for a kitchen manufacturer where direct sunlight pooling and high laminated board reflectivity is anticipated.

[0190] From the 3 spectrum channels for each matrix element, the controller is able to identify the relative brightness and also a probability of whether a high brightness area is resulting from sunlight pooling. Further analysis of contrast levels in that element will confirm this situation. By comparing the overall brightness and its current power level, it will decide whether to increase output to reduce contrast. If so, and after the adjustment, it will be able to measure whether contrast reduction has been achieved.

[0191] Figure 1 1 illustrates another embodiment, in which a collaborative, or group, of controllers, are able to have a controller react to occupancy detection events signalled by other controllers. This embodiment as application where behaviour differs by area, such as in industrial sites, car parks and project construction spaces.

[0192] The group requires no more than a single controller, through a wired or wireless medium, to signal to other controllers in its vicinity. As shown in Figure 1 1 , after controller 'A' detects a presence event (Step 1 in Figure 1 1 ), controller A transmits an encoded, undirected message that it has detected an initial occupancy event. External to the encoding, the notification includes a unique identifier code (Step 2). In modern wireless systems, that unique identifier could simply be the Controller's MAC address, an automatically generated identifier embedded in the communications protocol stack. When controller 'B' receives an arbitrary occupancy message (Step 3), that message is timestamped and recorded (Step 5). If controller 'B' then detects its own sensing of presence within a period of several seconds (Step 4), then controller 'B' will record the received event in its correlation algorithm (Step 5). Once the statistical correlation of its detection to that of another controller's detection has passed a threshold, it will start to react to that received event as if it was its own event.

[0193] The correlation function is divided into a long term and short term correlation storage where short term positive correlation relationships cause the long term correlation to slowly evolve (Step 6).

[0194] A further reinforcement of this relationship occurs when the receiving controller (controller 'B' in this example) detects an increased light level in its field of view, at the same time that it receives the detection signal. This detection will imply a close physical relationship between controller 'A' and controller 'B', so that the variable time response to the received detection will be short.

[0195] If there is no observable correlated light increase (allowing for ambient adjustment), the receiving controller (controller 'B' in this example) will calculate an appropriate time delay based on some period less than the average correlated delay times between the received signal and its own detection. [0196] A controller can store an evolving time delay period between when it receives a group message and when itself detects a presence event.

• In the case of a pallet rack aisle with guided forklifts, this value may be 3 seconds with a variance of 0.5;

• In that case of an open factory, this value may be highly variable but may average say 5 seconds with a variance of 3;

• In the case of an office hallway, it may be 2 seconds with a variance of 1 .

[0197] In the first case, because of the high probability and short time frame, the observing controller can choose to respond, with a rapid light increase, as soon as it senses and ambient light change in its field of view, as there is a high probability that it will detect occupancy in 2-3 seconds.

[0198] In the second and third cases, because of the higher relative percentage variance, the controller can choose to respond with a slowly increasing light output that ceases once the average plus variance time has been exceeded and it has not detected its own occupancy.

[0199] Figure 12 shows a specific example of Figure 1 1 , where now the group control learning is in regard to controlling a light (luminaire). The flow chart illustrates the communication occurring between the controllers in the group.

[0200] A further embodiment wherein a group response is obtained, is from a sudden brightening of lamp based illumination in a controller's view, where there is no local presence detection. This sudden brightening will activate the sensing controller. A situation where this could occur is on either side of a transparent partition (e.g. a lift foyer door, or glass office wall) through which a presence detection signal cannot pass. This can be a learned response from correlation of a brightness event, followed repeatedly by a (slightly delayed) detection event.

[0201] A further embodiment uses independent controllers that may not have direct control of any lamps but may be placed in locations to give early warning of an occupancy event. This invention allows such a controller to be placed anywhere, without any specific configuration, as the messaging receiving controllers will form their own correlation function against both a) controller T 's signalling and b) their local presence detection events. To minimise direct interference in this process either by inadvertent or malicious imitation of Group controllers, the transmitted event message is encoded with an algorithm that takes as an input the transmitter's MAC value to make a basic group correlation.

[0202] A further embodiment of the present invention involves customizing the behaviour of groups of controllers so as to be adapted for special environments. For example, in the lighting control domain, cases of this would include: multi storey environments where group related signals can penetrate multi-level buildings, or a wired network encompasses controllers on more than one floor; car parks that consist of multiple types of trafficable areas such as transit lanes, lobbies, parking bays and storage cages; long distance areas or areas of poor communication that require received messages to be relayed or redistributed; an "All Points" activation from a specific controller's presence detection; changed responses relating to a controller taking over the role of an emergency light controller in the event of power loss; a specific message issued by an external device that is reacted to as a building emergency response, such as triggering flashing patterns in particular lamps.

[0203] Each of these cases require a degree of group behaviour, however specific cases can be derived / supported from adaptive learning algorithms controlling the processor in the controller, whilst others will require specific configuration and in some situations special hardware.

[0204] A further embodiment of the present invention involves a security value being selected within a controller. When the security value is selected, many of the sensing control attributes may have similar settings to those of an interior building lamp controller. There are however, differences in the way ambient light sensors are acted on and also how its learning algorithms apply to sensory inputs. For example, it would be normal behaviour for the controller to not activate the security lamp until a minimum level of Lux input has been sensed, or a real-time clock notification has occurred, and then for it to become active at monitoring its surrounds. An enhancement to this security function, through detailed attribute profiling and adaptive learning by the capability of observing, storing and reacting to repetitive behaviour patterns. These learned behaviours include: the ability to coordinate lighting actions with other controllers based on group learning; the ability to time stamp and record for reporting events that do not follow normal patterned behaviour; the ability of video sensors to switch to an image capture mode, for later download reporting (or transmission for storage), when abnormal behaviour patterns are detected.

[0205] A further embodiment of the present invention relates to the use of environmental sensors, in areas other than occupancy and ambient light. This use allows the controller to produce command signals for other environmental systems. This function can be as simple as responding, via a network message from a building management system (BMS), for a sensor reading, through to directly controlling devices such as vents and dampers.

[0206] In this particular case, the adaptive learning system develops a measurement value that is more complex than a straight sensor reading, in that it is based on cross-referencing with other sensors such as activity level and historical patterns of previous actions re-response.

[0207] A further embodiment of the present invention involves the controller being able to directly interface to a very wide range of third party environmental system hardware both directly by itself, and also with the assistance of its IP65+ rated interconnect, shown in Figure 13. IP65 is an

Environmental Protection standard indicating that the device has Ingress Protection against dust and liquid (water). This module performs the following additional support functions: extreme power supply conditioning derived from the environmental system to supply the controller over and above what the controller itself is able to manage; provide an electrical disconnect function to enhance control protocols such as 1 -10V that are not able to be provided to it natively; electrical isolation of control signals where the controlled device cannot tolerate connected power supply and control circuits; provide multiple opportunities to support wiring gauge connection transformations from across the 28 to 12 AWG spectrum; provide a protected IP65+ environmental protection to all of the wiring feeds connectivity; provide an adjustable strapping field to support the full lifecycle of the controlled device from factory test through to final configuration and commissioning; provide alternate packaging options for integrated sensors as opposed to separate sensor pods.

[0208] In an embodiment a very detailed model of environmental control can be determined with a wide range of environmental attributes that a high functioning control system can consider in its determination of the most appropriate output control settings. A representative range of control attributes that this invention seeks to more effectively manage, through applying standardised Use Cases to a range of Occupancy usage types is below. Some of these attributes define how

Controllers, equipped with learning algorithms, are able to use this observed knowledge to modify both short-term and longer-term control behaviour, in response to changing environmental and usage patterns. 1.1 Presence or Occupancy Detection

1.1.1 Maximum Light Value

[0209] The allowable maximum value (%) of the light output. It could be de-rated because of driver -LED array mismatch or because of the Lux level requirement, from the lighting design.

1.1.2 Dwell Period

[0210] The time that a light stays on Maximum power; after the last Presence detection event it has detected directly, or has been conveyed to it.

1.1.3 Dimmed Value

[0211] The level of output that a light goes to initially after the Dwell period has elapsed.

1.1.4 Dimmed Period

[0212] The time that a light stays on its Dimmed value; after the last Presence detection event it has detected directly, or has been conveyed to it.

1.1.5 Dim Down Rate

[0213] The speed that the light output changes from one level state to the next lower level state.

1.1.6 Dim State Transition Rate

[0214] The speed that the light output changes from a dimmed state to the Maximum level state once a presence detection event has occurred.

1.1.7 Absolute Minimum Value

[0215] The minimum value that the light output can descend to. This may not be fully "Off" because of a number of reasons, including Driver limitations and customer preferences to not have a location go completely dark.

1.1.8 Device sensor Configuration and Control Attributes

[0216] There are numerous attributes related to Presence detection parametric that can have a direct effect on sensor performance. These include:

• Sensing threshold level (at what level is a sensing event recognised);

• Detection Sample and/or cycle time;

• Filter attributes;

• Sensing area; • Occupancy Sensing Size;

• Signal Detection Algorithm selection;

• Multi-sensor Correlation.

1.1.9 Controller 3D Co-ordinates

[0217] For explicitly defining a Controlling device's location in 3D space. This is particularly relevant in multi-storey establishment where signals travel between floors.

1.1.10 Device Comparison Algorithm

[0218] Which rule numbers is the controller to use in order to apply to the ID of a transmitting controller determine whether the activity of another controller should cause it to take action and what action that should be.

1.1.11 Device Comparison Rule 1 -n

[0219] These rules define pattern matching masks that are applied using the device comparison algorithm. It could be a decision whether to on forward a message with or without amendment.

1.1.12 Occupancy Type Detection and Rules selection

[0220] The type and relative importance correlation rules for Controllers that are equipped with more than one sensor that is capable of generating occupancy information.

1.1.13 Device Behaviour on Comparison

[0221] Selection of which rules and what control protocol should be actioned to generate a control output, when a positive detection event is processed.

1.1.14 Device Onward Signalling Algorithm

[0222] A set of rules for determining what, if any, on forwarding of signalling information and what fields, if any, should the on forwarding Controller substitute.

1.2 Ambient Light, Temperature and Air Quality Control

1.2.1 Light Harvest Baseline LUX

[0223] A level of perceived ambient Lux; above which a controller begins to reduce the power of an associated luminaire. 1.2.2 Reflective Light Sensitivity

[0224] Because the position that a controller's light sensor differs from that of a person at ground level, a compensating factor needs to be built into its calculation. This factor takes into account floor reflectivity, field of view and controller elevation.

1.2.3 Light Power Reduction rate

[0225] Because of the complex and unpredictable feedback factors at work with multiple luminaires with overlapping lighting patterns, as depicted in Figure 3, it is not possible, and even preferable, for a controller to attempt to exactly track changes in natural light illumination. One reason why its response should be less that of a 100% compensatory response is the need to minimise contrast in its field of view between naturally lit areas and ones in shadow. It has therefore been found that a fixed power reduction rate per extra measured lumen allows for the subjective variations across differing work spaces.

1.2.4 Ambient Light Adjustment Frequency

[0226] This attribute allows for an averaging process and effectively setting a low pass filter in the luminaire feedback loop so that any transient event will not trigger any visually distracting change of luminaire output

1.2.5 Ambient Calibration Control

[0227] This invention describes mechanisms to allow a controller to evaluate its own and natural contributions to overall light level and this attribute provides the configuration values to drive self- calibration.

1.2.6 Ambient Sector Quantity and Location

[0228] For ambient sensing coming all or in part from a video sensor, this attribute specifies which parts of its field of view should be used for ambient sensing.

1.2.7 Ambient Sector Correlation

[0229] This attribute controls the relationship between a natural light and overall level sensor so that a better-informed decision can be made as to what amount of power the controller should adjust the luminaire output by. 1.2.8 Minimum Ambient Level

[0230] In many operational spaces users prefer that there is always a minimum level of illumination coming from the luminaire. This attribute defines what the default value of this minimum should be.

1.2.9 Ambient Type Detection and Rules selection

[0231] The type and relative importance correlation rules for controllers that are equipped with more than one sensor that is capable of generating environmental level information.

1.2.10 Lighting Colour

[0232] It is well known that human mood is affected by illumination's colour temperature and with increasing use of control systems such as enhanced DALI commands that can control colour change. This feature needs to be formally defined with control attributes. In an embodiment of the invention, the colour attribute is able to be allocated a value for each defined time period.

1.2.11 Time of Day /Week

[0233] There is a need for controllers that operate on a defined multi-shift or 24x7 cycle to have the ability to alter key attributes depending on what phase of the repetitive cycle the controller is currently operating. One such arrangement is that the weekly cycle is split into 3 time bands, Primary, Shoulder and Off-peak activity.

1.2.12 Temperature Control

[0234] Attributes of temperature control are very similar to luminaire control - particularly if temperature changes can be influence by many decentralised ventilation, heat sources or split system units.

1.2.13 Airflow, Humidity and Gas Build-up Control

[0235] Attributes of ventilation control can be used to adjust the influence of sensors across these general domains.

1.2.14 Sound Control

[0236] Sound sensors attribute control can control the influence such measurements can have on:

• Occupancy detection;

• Ventilation Control (from external noise);

• Measurement and Reporting of out of allowable ranges. 1.2.15 Weather Forecast Input Attributes

[0237] The ability to take in forward estimates of temperature / humidity profiles can be controlled by attributes related to relevant actionable control outputs - such as pre-warming, pre-ventilation, etc.

1.3 Security and Management

[0238] For a controller to perform security related functions, this needs to be specified in the sensor configuration table.

1.3.1 Device Hardware Configuration

[0239] The hardware configuration table defines the full set of sensor configurations, as well as certain, specialised output protocol variants (e.g. DSI, PWM).

1.3.2 Access Control

[0240] The access control specifies the level of encoding and the level of access allowable for remote configurability.

1.3.3 Access Cut-off Time

[0241] The most secure method of remote configuration access security is to prevent access completely. However, remote access may be necessary for commissioning purposes and rearranging work areas. To support these opposing requirements, the controller has an attribute that specifies a period after which remote access will not be able to occur.

1.3.4 Access Security Control

[0242] This time period can be reset by, explicit actions such as power cycling or by receiving an encoded message from a trusted agent.

1.3.5 Reporting Profile and Distribution

[0243] Any storage of operational and sensor data for reporting purposes and subsequent interrogation and/or downloading needs to defined in the reporting control table attributes.

1.4 Existing stored program systems

[0244] A good example of the current state of the art for programmable control of lighting can be found in the DALI protocol use of "scenes"; whereby a stored sequence of "scenes" can cause a particular dimming sequence to be selected and this dimming sequence contains the from-to values and the dimming rate. [0245] Another example of stored control sequences can be found in stage lighting apparatus where entire groups of lights are controlled in sequence to perform complex intensity colour, focus and directionality adjustments.

1.5 Reduced Set of Controllable Attributes

[0246] Most Presence, Temperature and Ambient Light Controllers that are fitted to individual environmental controlled units (Lights, ventilation fans, movable skylights, split systems, etc.) rely on a very small set of control attributes and, therefore, the sophistication of their control response is very limited.

[0247] For luminaires, the Ambient Control attribute set is commonly restricted to just one value - the threshold at which a lamp either switches off completely or dims down proportionally at some fixed rate. All of the additional control attributes, abled to be profiled, as outlined in above are not implemented, or are hard coded

[0248] For occupancy settings, control attributes are usually limited to 1 or 2 timeout values with other control attributes not implemented or having hard coded values.

[0249] For other environmental sensing such as temperature, many controllers are limited to On or Off with some having a proportional response related to room temperature difference from the set point.

1.6 Mechanical Strapping Fields

[0250] In this type of controller, the actual setting levels are selected by a jumper field arrangement of shorting straps. As each of these strapping points needs to be able to be independently "read" by the controller, the quantity of strapping points, and hence the attribute range and control scope and resolution of this technique is severely limited.

1.7 Analogue potentiometers

[0251] Some basic occupancy and ambient controls use uncalibrated analogue potentiometers to set ambient threshold, presence detection thresholds and light time. This type of attribute setting lacks any level of precision and severely limits the number of controllable attributes. 1.8 Miniature DIP Switches

[0252] There exist sensors for the commercial / industrial markets that utilise a series of dual in-line package (DIP) switches. Although slightly more sophisticated than strapping fields and

potentiometers, they also suffer seriously from limitations on the available resolution of settings and the number of attributes that can be specified.

1.9 Stored values by Remote control

[0253] A number of wireless transmission means have been used to remotely set attributes in the non-volatile memory of environmental controllers. These include: -

• Proprietary infrared;

• Proprietary Wi-Fi;

• Various proprietary protocols via a fixed LAN;

• ZigBee and Bluetooth V4 & 5. Both of these communications stack architectures

standardise, to some degree, classification of the service attributes required for multi-vendor interoperability, although this level of specification does not cater for the majority of attributes identified as needed definition and support within this invention.

[0254] Modifications may be made to the present invention within the context of that described and shown in the drawings. Such modifications are intended to form part of the invention described in this specification.

Claims

1 . A method for controlling an electrical energy consuming device, comprising the steps of: determining a plurality of profiles for a plurality of types of space within which the electrical energy consuming device is able to be used, where each profile defines a set of control attributes for a corresponding type of space;

selecting one of the profiles;

receiving an input from the sensor;

2. A method according to claim 1 , wherein the signal output from the controller is for controlling light output from the electrical energy consuming device.

3. A method according to claim 1 , wherein the signal output from the controller is for controlling heat output from the electrical energy consuming device.

4. A method according to claim 1 , wherein the method further comprises preconfiguring the controller with the plurality of profiles.

5. A method according to claim 4, wherein the selection of one of the profiles comprises preselection of one of the profiles as a default.

6. A method according to claim 4 or 5, wherein the selection of one of the profiles comprises receiving a command from an external device.

7. A method according to any one of the previous claims, wherein the sensor is at least one of: a thermal sensor, an electromagnetic sensor, a mechanical sensor, a chemical sensor, an optical radiation sensor, an ionizing radiation sensor, a biological sensor, a geodetic sensor, a light sensor, a force sensor, a humidity sensor, an air quality sensor, a camera, a payment sensor, a microphone, a Radio Frequency Identification (RFID) reader, a Hall Effect sensor, and a limit-switch.

8. A method according to any one of the previous claims, wherein the method further comprises creation of an electrical energy consuming device layout plot to classify the area into different types of space where each type of space has a different usage for the electrical energy consuming device(s) and the associated controller(s) within the space.

9. A method according to claim 8, wherein, the layout plot provides an electrical energy consuming device installer with details of the type of electrical energy consuming device to install; where to install the determined type of electrical energy consuming device; the type of profile to select for the electrical energy consuming device and an identifier of its associated controller; and a map to record details of each electrical energy consuming device and the identifier of its associated controller.

10. A method according to any one of the previous claims, wherein the profile is selected according to the type of electrical energy consuming device being controlled, the type of space in which the electrical energy consuming device is to be used and the type of usage of the space.

1 1 . A method according to any one of the previous claims, wherein each profile comprises a plurality of control attributes, where each control attribute is suitable for configuring an element of a control paradigm for the electrical energy consuming device.

12. A method according to any one of the previous claims, wherein selection of the profile comprises monitoring the sensor input over time and changing the selection of the profile according to patterns recognised in the monitored sensor input.

13. A method according to any one of the previous claims, wherein the method further comprises adapting the default profile based on a recording of inputs received from the sensor over time by use of a learning algorithm.

14. A method according to any one of the previous claims, wherein the method further comprises autonomous adaptation of the controller's behaviour via the default usage class configuration of the controller including observational and learning algorithms in the event of an electrical energy consuming device being present in a space that does not have a predefined usage class associated with the space for the controller.

15. A method according to any one of the previous claims, wherein the controller creates a new profile from an already selected profile, by adapting the selected profile using machine learning, so as to create a new profile that matches the new environmental situation.

16. A method according to claim 15, wherein the new profile is determined by analysing stored sensor inputs and fitting a model to these inputs in correlation with the stored profiles.

17. A method according to claim 15 or 16, wherein the newly created profile is stored in a profile list table for later matching so that future environments with the same requirements can utilise the newly created profile.

18. A method according to any one of the previous claims, wherein the sensor comprises a presence detection sensor, wherein the presence detection sensor can be refined so as to increase the detection sensitivity of the sensor.

19. A method according to any one of the previous claims, wherein the sensor comprises a light sensor wherein the light sensor comprises at least one of a human eye perceivable light response sensor and a colour camera element light sensor.

20. A method according to any one of the previous claims, wherein the profile comprises at least one attribute which can be modified, wherein the at least one attribute includes at least one of: maximum light value, dwell period, dimmed value, dimmed period, dim down rate, dim state transition rate, absolute minimum value of output, sensing threshold level, detection sample time, detection cycle time, filter parameters, sensing area, occupancy sensing size, signal detection algorithm selection, multi-sensor correlation, controller 3D coordinates, device comparison algorithm selection, device comparison rule 1 -n, occupancy type detection and rules selection, device behaviour on comparison selection, device onward signalling algorithm selection, light harvest baseline Lux, reflective light sensitivity, light power reduction rate, ambient light adjustment frequency, ambient calibration control, ambient sector quantity and location, ambient sector correlation, minimum ambient level, ambient type detection and rules section, lighting colour, time of day, time of week, temperature, air flow, humidity, gas, sound, weather forecasts, security access levels and access cut-off periods.

21 . A device for controlling an electrical energy consuming device, comprising:

at least one sensor for sensing an aspect of a region;

a controller for receiving a plurality of profiles and an input from the sensor; wherein the controller operates according to a selected one of the profiles, or a self-created profile, and the input from the sensor;

22. A device according to claim 21 , wherein the controller comprises a processor for creating the self-created profile.

23. A device for controlling an electrical energy consuming device, comprising:

at least one sensor for sensing an aspect of a region;

a controller for receiving an input from the sensor;

24. A system for controlling a plurality of electrical energy consuming devices, comprising: a plurality of sensors, each sensor for sensing a respective region within the sensor's range when installed in an area which comprises the respective regions;

wherein each controller receives an input from the associated sensor;

25. A system for controlling a plurality of electrical energy consuming devices, comprising: a plurality of sensors, each sensor for sensing a respective region within the sensor's range when installed in an area which comprises the respective regions;

wherein each controller operates according to the automatically generated profile so as to output a signal to control the respective electrical energy consuming device.

26. A method for controlling a plurality of electrical energy consuming devices, comprising the steps of: sensing a respective region determined by a respective sensor of a plurality of sensors, wherein the respective region is within the respective sensor's range when installed in an area which comprises the respective regions;

27. A luminaire controller comprising:

a processor configured to process the video images for detecting movement;

28. A luminaire controller according to claim 27, wherein the processor is adapted to de-focus on or decrease sensitivity on areas where movement is not previously detected.

29. A luminaire controller according to claim 27 or 28, wherein the processor is adapted to de- focus on or decrease sensitivity on areas where there is repetitive high frequency movement.

30. A method of controlling a luminaire, comprising:

measuring illumination of an area illuminated by the luminaire;

capturing video images of the area illuminated by the luminaire; and

processing the video images for detecting movement;

storing image areas where a higher frequency of movement is detected over time;

31 A luminaire controller comprising: a light detector for measuring illumination of an area illuminated by a luminaire;

a processor for controlling an output signal for controlling the luminaire;

32. A luminaire controller according to claim 31 , wherein the brief reduction in luminaire output is repeated so as to establish over time the luminaire output required to achieve a desired overall lux in the area.

33. A luminaire controller according to claim 32, wherein the desired overall lux is the maximum output of the luminaire in conditions of no ambient light.

34. A luminaire controller according to any one of claims 31 to 33, wherein the processor is configured to make small up or down adjustments in luminaire output over a longer period, and to measure the response received by the ambient light detector so as to determine a response by other luminaires that produce light detectable by the ambient light detector.

35. A luminaire controller according to any one of claims 31 to 34, wherein the luminaire controller comprises a visible light sensor and a near-infrared sensor and is configured to determine what ambient light is from sunlight and what ambient light is from other sources based on measurements of the visible light and near-infrared sensors.

36. A method of controlling a luminaire controller comprising:

measuring illumination of an area illuminated by the luminaire;

controlling an output signal for controlling the luminaire;

briefly reducing the luminaire output;

measuring the corresponding brief reduction in reflected light received;

calculating the required output control to the luminaire based on the measured reduction in the reflected light.

37. A luminaire controller comprising:

a colour camera for receiving an image of an area illuminated by a luminaire;

a processor for controlling an output signal for controlling a luminaire;

wherein the processor reads an automatic brightness control value of the camera and determines the ambient light level based on the brightness control value; wherein the processor derives spectral components average levels from the camera and analyses the spatial area for those areas that are particularly bright.

38. A luminaire controller according to claim 37, wherein the analysis comprises comparing the level of red to green and blue in order to differentiate the relative sunlight to luminaire ratio.

39. A luminaire controller according to claim 37 or 38, wherein the processor is configured to momentary change the colour output of the luminaire and to process the measured change in overall spectral levels in order to determine the appropriate luminaire output by compensating for contrast between bright and dark in the brightness control value.

40. A method of controlling a luminaire comprising:

receiving an image of an area illuminated by the luminaire with a camera;

controlling an output signal for controlling the luminaire;

41 . A luminaire controller comprising:

an image sensor for receiving an image of an area illuminated by a luminaire;

wherein the processor is configured to:

42. A luminaire controller according to claim 41 , wherein the processor is further configured to determined responses of other luminaires to the changes to the output.

43. A luminaire controller according to claim 41 or 42, wherein the processor is further configured to determine a period of time over which the determinations are made.

44. A luminaire controller according to any one of claims 41 to 43, wherein the processor is further configured to determine a number of samples to take in order to make the determinations in the determined period of time.

45. A method of controlling a luminaire comprising:

receiving an image of an area illuminated by the luminaire;

46. A luminaire controller comprising:

an image sensor for receiving images of an area illuminated by a luminaire;

wherein the processor is configured to use a statistical model to:

47. A method of controlling a luminaire comprising:

receiving images of an area illuminated by the luminaire;

using use a statistical model to: decide the how the image area is divided into analysis sub-area based on common levels of image activity;

48. A luminaire controller comprising:

a processor for controlling an output signal for controlling the luminaire;

49. A method of controlling a luminaire comprising:

measuring illumination of an area illuminated by the luminaire;

outputting a signal for controlling the luminaire;

nulling out a contribution of natural light and other lighting devices;

50. A luminaire controller comprising:

51. A method of controlling a luminaire comprising:

evaluating the changes in output of the spectrum channels of the image sensor;