WO2018149901A1

WO2018149901A1 - Route planning of a vessel

Info

Publication number: WO2018149901A1
Application number: PCT/EP2018/053737
Authority: WO
Inventors: Juha Rokka; Kenneth SOLBERG; Oskar Levander; Krzysztof SWIDER
Original assignee: Rolls Royce Oy AB
Current assignee: Kongsberg Maritime Finland Oy
Priority date: 2017-02-16
Filing date: 2018-02-15
Publication date: 2018-08-23
Anticipated expiration: 2019-08-16
Also published as: FI20175137L; FI20175137A7

Abstract

It is an object to provide route planning of a vessel. In an embodiment, an apparatus is disclosed. The apparatus comprises: a processor; and a storage comprising a set of instructions that, when executed by the processor, cause the apparatus to: receive first data regarding reliability of a first wireless connection; receive second data regarding cost characteristics of a first wireless connection; receive third data regarding reliability of a second wireless connection; receive fourth data regarding cost characteristics of a second wireless connection; receive a route plan of a vessel configured to navigate and control the vessel, wherein the first and the second wireless connections are used to control the route plan and navigate the vessel;and modify the route plan based on the first, second, third and fourth data so as to process the reliability and cost characteristics of the wireless connections for the route plan. The route of the vessel may be optimized and balanced with respect to safety of the connections and cost of the connections. Other embodiments relate to a ship or a shore operator having the apparatus, and a method and a computer program for performing the operations of the apparatus.

Description

ROUTE PLANNING OF A VESSEL

BACKGROUND A vessel, such as a ship, may detect its en^¬ vironment and surroundings by various detectors which may be alternatively referred to as sensors. The de^¬ tectors provide information to a control system of the vessel, which may be used to render data information about the current situation of the vessel or an object nearby the vessel. This data information may be out^¬ put, for example for control, navigation or monitoring purposes of the vessel. SUMMARY

This summary is provided to introduce a se^¬ lection of concepts in a simplified form that are fur^¬ ther described below in the detailed description. This summary is not intended to identify key features or essential features of the claimed subject matter, nor is it intended to be used to limit the scope of the claimed subject matter.

It is an object to provide route planning of a vessel. The object is achieved by the features of the independent claims. Further embodiments are de^¬ scribed in the dependent claims. In an embodiment, an apparatus is disclosed. The apparatus comprises: a processor; and a storage comprising a set of instruc^¬ tions that, when executed by the processor, cause the apparatus to: receive first data regarding reliability of a first wireless connection; receive second data regarding cost characteristics of a first wireless connection; receive third data regarding reliability of a second wireless connection; receive fourth data regarding cost characteristics of a second wireless connection; receive a route plan of a vessel config^¬ ured to navigate and control the vessel, wherein the first and the second wireless connections are used to control the route plan and navigate the vessel; and modify the route plan based on the first, second, third and fourth data so as to process the reliability and cost characteristics of the wireless connections for the route plan. The route of the vessel may be op^¬ timized and balanced with respect to safety of the connections and cost of the connections.

According to an embodiment, the set of in^¬ structions, when executed by the processor, further cause the apparatus to: estimate the reliability and cost characteristics of the first and the second wire^¬ less connections, wherein the route plan is based on the estimate. The whole intended route may be de^¬ signed .

According to an embodiment, the reliability of the wireless connection comprises a degree of prop^¬ erties with respect to delivery of data to the intend- ed recipient; or a probability level indicating how reliably the wireless connections can provide data communications. Different kinds of levels may be used for categorizing the reliability of the connection for optimizing the route.

According to an embodiment, the set of in^¬ structions, when executed by the processor, further cause the apparatus to: determine a cell based network for determining the route, wherein cells indicate the reliability of the wireless connections. A network with cells facilitates optimization of the route.

According to an embodiment, the set of in- structions, when executed by the processor, further cause the apparatus to: determine an estimate for the first data and the third data; compare the estimate to the first and the third data so as to assess the route plan. A relationship between safety and estimated re- liability of the route may be determined. Safety and predicted reliability can be linked together such that the operator may continuously be able to assess wheth^¬ er the safety is at risk, if the predicted reliability is dropping to low level.

According to an embodiment, the set of in^¬ structions, when executed by the processor, further cause the apparatus to: plan and schedule transmission of large amounts of data. Consequently, the transmis^¬ sion cost can be minimized.

According to an embodiment, the reliability is based on one of the following characteristics of the wireless connection: link redundancy, coverage, capability such as throughput or latency, age of meas^¬ urements, signal strength, weather, moving objects with a certain height and ability to navigate around them, or stability. Existing network parameters for the connection may be used for determining the route.

According to an embodiment, the cost charac^¬ teristics are based on one of the following: a func- tion of an expected amount of data to be transferred, connectivity technology used, or cost of a fallback to another wireless connection. The cost may be consid^¬ ered for the route optimization.

According to an embodiment, the set of in^¬ structions, when executed by the processor, further cause the apparatus to: navigate the vessel according to the modified route plan. The vessel may be autono^¬ mously navigated via the optimized route.

According to an embodiment, the first wire^¬ less connection comprises mobile communication. It provides relatively good capacity with reasonable costs .

According to an embodiment, the second wire^¬ less connection comprises satellite communication. Satellite communication may be widely available.

According to an embodiment, further included is a third wireless connection, which comprises Wi-Fi communication; wherein the set of instructions , when executed by the processor, cause the apparatus to: re^¬ ceive fifth data regarding reliability of the third wireless connection; receive sixth data regarding cost characteristics of the third wireless connection; wherein the third wireless connection is further used to control the route plan and navigate the vessel; and modify the route plan further based on the fifth and sixth data so as to process the reliability and cost characteristics of the wireless connections for the route plan. The WiFi connection may be used to further optimize the route.

Other embodiments relate to a ship comprising the apparatus or a ship operator on a shore comprising the apparatus . According to another embodiment, a method comprises: receiving first data regarding reliability of a first wireless connection; receiving second data regarding cost characteristics of a first wireless connection; receiving third data regarding reliability of a second wireless connection; receiving fourth data regarding cost characteristics of a second wireless connection; receiving a route plan of a vessel configured to navigate and control the vessel, wherein the first and the second wireless connections are used to control the route plan and navigate the vessel; and modifying the route plan based on the first, second, third and fourth data so as to process the reliability and cost characteristics of the wireless connections for the route plan.

According to an embodiment, a computer program comprises a set of instructions which are config^¬ ured to cause a computer to perform the steps of the method when executed.

Many of the attendant features will be more readily appreciated as they become better understood by reference to the following detailed description considered in connection with the accompanying drawings .

BRIEF DESCRIPTION OF THE DRAWINGS

The present description will be better understood from the following detailed description read in light of the accompanying drawings, wherein:

FIG. 1 illustrates a schematic view of a compu^¬ ting apparatus receiving various inputs and con^¬ trolling different data layers for determining a navigation route of an autonomous vessel accord^¬ ing to an embodiment;

FIG. 2 illustrates a schematic top view of a nav^¬ igation route of an autonomous vessel in relation to different kinds of connectivity characteris^¬ tics according to an embodiment;

FIG. 3 illustrates a schematic view of redundant network capabilities for a navigation route of an autonomous vessel according to an embodiment;

FIG. 4 illustrates a schematic view of a connec^¬ tivity network grid with respect to the redundant capabilities according to an embodiment;

FIG. 5 illustrates a block diagram of a computer for configuring a navigation route of an autonomous vessel according to an illustrative embodi^¬ ment; and

FIGS 6, 7 and 8 illustrate examples of a hidden Markov model usage for determining the route pre^¬ diction .

Like reference numerals are used to designate ke parts in the accompanying drawings.

DETAILED DESCRIPTION

The detailed description provided below in connection with the appended drawings is intended as a description of the present embodiments and is not in^¬ tended to represent the only forms in which the pre^¬ sent embodiments may be constructed or utilized. How^¬ ever, the same or equivalent functions and sequences may be accomplished by different embodiments.

Although the present embodiments may be de^¬ scribed and illustrated herein as being implemented in a single ship, this is only an embodiment of a vessel having a computer with connectivity and not a limita^¬ tion. As those skilled in the art will appreciate, the present embodiments are suitable for application in a variety of different types of systems and vessels, for example in a single ship, many ships, a maritime sys^¬ tem, a decision support tool for a user or crew, a marine remote control system, an autonomous marine navi^¬ gation or position system, or other marine systems for autonomous navigation and route planning. The vehicle may be other kind of moving object than a ship, for example a truck or an airplane.

Although the present embodiments may be de^¬ scribed and illustrated herein as being implemented in autonomous vessels, this is only an embodiment of a vessel having a computer with connectivity and not a limitation. As those skilled in the art will appreci^¬ ate, the present embodiments are also suitable for ap^¬ plication in remotely operated vessels and/or any com- binations of the autonomous and remotely operated ves^¬ sels.

An embodiment relates to a vessel being au^¬ tonomously navigated. The navigation is based on sev^¬ eral wireless connections which are transmitting data between the vessel and external systems. The external systems are required to configure the navigation route depending on various factors. Route planning is used to ensure that remote and autonomous navigation opera^¬ tion of the vessel can be optimized. This may be per- formed by supporting criteria such that a minimum re^¬ quirement of safety is reached with a given cost for the connectivity being applicable. Safety may depend upon connectivity, which is by its nature unreliable. The cost characteristics may also depend upon connec^¬ tivity, and if one connectivity link fails, it may fall back to a link which has a different cost. For example, it may be more expensive to use. Consequent^¬ ly, the probability of these two factors may change in time, and this is tracked and estimated. This can be estimated by using previously captured data from vari^¬ ous systems to indicate an intended route which com- plies with the requirements of cost and safety (both related to the applicable connectivity) . The route may be configured based on these requirements.

It should be noted that there may be one or many wireless connections which are used simultaneous- ly and are prioritized based on cost. Data may be dis^¬ tributed and weighted over different groups of connec^¬ tions with same cost or coverage.

FIG. 1 illustrates a schematic view of a com^¬ puting apparatus receiving various inputs (100,101,102) and controlling different data layers (103,104,105,106) for determining a navigation route of an autonomous vessel according to an embodiment.

The computing apparatus is connected to vari^¬ ous systems 100, 101 and 102, where the respective systems are responsible of updating data layers of in^¬ formation. The computing apparatus is connected to various different connections by a connectivity system 100. The connectivity system 100 uploads information relating to a Wi-Fi map layer, cellular map layer and satellite map layer. These layers illustrate the con^¬ nectivity and coverage of these networks. These layers have also information about characteristics of the re- spective connectivity. An SA system 101 updates ob^¬ jects both in a Lidar layer (short range) and a radar layer. The objects therein may be physical obstacles, for example other vessels, landmarks, harbor construc- tions, islands, etc. A weather information system 102 is configured to update a weather layer.

The systems 100, 101, and 102 are maintaining their own data layers comprising connection parameters and spatial information represented in these layers. Probability information may also be maintained in these systems 101, 102, and 103 so that they have an estimation of upcoming changes. An optimal route can be made of filters such as costs and a minimum re^¬ quired probability of achieving adequate coverage with certain communication capabilities (for example throughput and latency) in order to process the level of a tolerable risk for safety. A spatial analysis al^¬ gorithm of the computing apparatus may be used to pro^¬ cess the layers and create the route with certain cri- teria.

For example, the connectivity system 100 up^¬ dates one layer with spatial coverage for each connec^¬ tivity with associated parameters: cost, link redun^¬ dancy, time to measure, probability of connectivity with latency and coverage (redundant links measured packet loss, signal strength, impact of weather and line of sight, stability of the links, etc.), proba^¬ bility of costs (for example, a failover results in a different cost scheme) , throughput and latency, net- work load.

A function of the probability of safety of the connection may depend on the following parameters: A function of link redundancy, coverage, capability (such as throughput, latency) , SLA, age of measure^¬ ments, signal strength, weather, moving objects with a certain height and ability to navigate around them, stability) .

A function of the probability of cost may de^¬ pend on the following parameters: A function of the expected amount of data to transfer, connectivity technology used and cost of an eventual fallback to another technology, SLA.

The different map layers of the systems 100, 101 and 102 are used to ensure that criteria are met based on these functions. According to an embodiment, the computing apparatus may be located in an operator of the autonomous vessel on shore. According to anoth^¬ er embodiment, the computing apparatus is on the ves^¬ sel .

FIG. 2 illustrates a schematic top view of a navigation route 108, 109 of an autonomous vessel 107 in relation to different kinds of connectivity accord^¬ ing to an embodiment. The route may be determined as described in various embodiments.

The vessel 107 is autonomously navigated on an original route 108. The computing apparatus as de- scribed in the embodiments is determining the route. The computing apparatus changes route 108 to route 109. The new route 109 travels through cellular con^¬ nections that are established by a cellular network 110 having base stations 110_1 - 110_5. Consequently, costs may be better optimized when the vessel 107 travels more within the coverage of the cellular net^¬ work. Furthermore, the redundancy of the connection may be improved since there are at least two connec^¬ tion options available: by satellite and by cellular connection .

According to an embodiment, the computing ap- paratus may be configured as follows. A given route shall not take place if the probability of the connec^¬ tivity capability does not reach a certain level. This can, for example, mean that the satellite connection cannot guarantee the bandwidth at a certain time or that the local cellular network is congested at some specific peak hours. This may also mean that the oper^¬ ation requires too much data to be sent to shore, such that it will become too expensive with the satellite connection .

An embodiment optimizes the route planning according to the costs of the connections. The satel^¬ lite communication cost may be comparable to the fuel cost, even exceeding it, if wideband data transmission is needed for real-time video delivery, for instance. Such a situation may arise if an unmanned ship has a problem and needs to be controlled remotely, or there is another reason for large data traffic. Terrestrial cellular communication is relatively cheap. Conse^¬ quently, significant cost savings can be achieved by changing the course of the vessel 107 through an area with cellular coverage.

For example, the fuel cost of a typical cargo ship may be somewhere in the range of 10-50 €/min. The satellite data transmission cost may be approximately 10-300 €/min for live video feed. The operational range is very wide as the requirements for video reso^¬ lution and frame rate may change quite a lot during remote operation. Most of the time a frame rate of 0.2 frames/s may be sufficient, while at times 20 frames/s may be occasionally needed. Typically, 720p or 1920p may be suitable frame sizes.

FIG. 3 illustrates a schematic view of redun^¬ dant network capabilities for a navigation route of an autonomous vessel according to an embodiment. The com^¬ puting apparatus may determine the route based on the redundant network capabilities.

Connectivity over open and unreliable net^¬ works may be considered as a commonly known problem with respect to reliability and single point of fail^¬ ure. By measuring the network characteristics (for ex^¬ ample delay and throughput) , the computing apparatus indicates the minimum redundant capability of the net^¬ work. This is beneficial in an operational context, as predictability and reliability are considered as some of the major premises of safe operations for route planning of the autonomous vessel 107. Whenever the minimum required capability is not satisfied, a single point of failure may break the route and operation, which again will have a negative impact to the opera^¬ tion. By detecting the performance of a connection, the computing apparatus may compare it with other con- nections to ensure that a minimum of two separate con^¬ nections provide the same delay and throughput.

One can always increase the amount of connec^¬ tions; however, the capability is about being able to control and guarantee that one connection provides the same capacity as another connection within a given area . The system makes it possible to extend the redundancy of the control system onboard a vessel 107 all the way to the operator station on shore, since it provides information about the level of redundancy all the way between the vessel 107 and the shore.

Referring to FIG. 3, an area 111_112 where cell 111 depicting a cellular connection overlaps with cell 112 depicting a WiFI connection gives a redundant network capability, for example within a data rate of 500-900 kbps and latency of 1-400 ms . A failover with^¬ in this area 111_112 will leave the operator on shore unnoticed. If the required bandwidth is beyond the ca^¬ pability of area 111_112, it will have an impact on the route, and this will not be a transparent failo- ver. The satellite connection 113 provides different kinds of capabilities, and it may not be able to pro^¬ vide a failover without degrading the route and opera^¬ tion.

According to an embodiment, there may be dif- ferent kinds of cellular or mobile networks, such as GSM up to the LTE based networks, each for a different wireless connection. Furthermore, instead of WiFi a different kind of short range wireless connection such as WiMax may be used. There may also be different sub- scriptions from different service providers, for each connection, cellular, short-range and satellite, etc. For example, five cellular connections from five dif^¬ ferent service providers with different capabilities.

FIG. 4 illustrates a schematic view of a con- nectivity network grid with respect to the redundant capabilities according to an embodiment. FIG. 4 shows network capabilities in different cells for route planning. Continuous monitoring of the connectivity may ensure the redundant network capabilities. The connectivity may be achieved by any connection tech^¬ nology with its limits and strengths, for example con- nections of the system 100 as illustrated in FIG. 1.

The computing apparatus uses information gathered from the vessels and various services on shore to decide if an operation has the minimum re^¬ quired network capability redundancy to perform an op- eration in a given geographical area. This may be in^¬ fluenced by:

the available connection characteristics in the particular area measured by the vessel;

the ability to guarantee a given network per- formance by a given supplier;

the ability of a supplier to indicate if a given network performance cannot be met due to capaci^¬ ty limitations;

the ability to use environmental information to indicate if a given connectivity technology may not be able to deliver a given performance in a given area .

The computing apparatus uses this information to decide whether the route and navigation operation may be performed in a given area at a given time, giv^¬ en all the uncertainties with poor and unreliable con^¬ nections .

In FIG. 4, reference 114 as indicated by a ^λ+' sign denotes a network cell having a satisfying level of minimum required connection capabilities for the route of the vessel 107. Thus, the minimum re^¬ quired redundant capabilities have been detected for the route of the vessel 107 in this cell. Reference 115 as indicated by a ^λ-^λ sign indicates that the con^¬ nectivity is redundant. There is sufficient network performance to perform the route. However, there is no extra redundant network capability. Reference 116 as indicated by a V sign indicates redundant connectiv^¬ ity. Consequently, the connectivity is not reasonable, or it is too bad for the required route. Reference 117 as indicated by a ^λ*' sign illustrates that there is no redundant connectivity. No route should be planned for these cells. Blank cells as illustrated by refer^¬ ence 118 indicate that no information about the con^¬ nectivity is available. No route should be planned for these cells either. In FIG. 4 the vessel 107 is trav- elling from cell 115 to cell 114, and consequently the connectivity improves. Cell 114 has a satisfying level of required capability.

The measurement results or other grid updates may be shared with other vessels. The connection net- work grid, such as the connectivity map, may be updat^¬ ed, even up to all the time, in several ways. Updates of the map may be downloaded from a server. Typically, some initial, estimated information is available for most relevant cells although the network properties in each cell may not have been measured. The vessel 107 can measure the signal strength of the communication systems and update the grid accordingly. The vessel 107 may upload the updates to the server in order to make them available for downloading by other vessels.

FIG. 5 illustrates an embodiment of compo^¬ nents of a computer 30, which may be implemented in any form of a computing and/or electronic device con- figured for performing the functionalities and opera^¬ tions of the embodiments of FIG. 1 to 4. They may re^¬ late to the computer 30 operating for determining the navigation route 107,108 for the autonomous vessel 107. The computing apparatus may be an embodied com^¬ puter 30. The computer 30 is equipped with a processor 36 and storage 37 comprising a set of instructions 38. The one or more processors 36 may be microprocessors, controllers or any other suitable type of processors for processing computer executable instructions to control the operation of the computer 30. The set of instructions 38 may comprise, for example, application software and platform software, such as an operating system to enable application software to be executed on the computer 30. When the set of instructions 38 is executed by the processor 36, the computer 30 is con^¬ figured to perform the functions and operations de^¬ scribed above in the embodiments of FIG. 1 to 4. Ac^¬ cording to an embodiment, the computer 30 is config- ured to receive, from the systems 100,101,102, infor^¬ mation relating to the route and connectivity; deter^¬ mine the route; change the route according to infor^¬ mation received from the systems 100,101,102.

The computer 30 is configured to predict the connection. The prediction is performed based on available connections with respect to cost and safety of the connections.

According to an embodiment, the cost of each connection is known, and each connection can have a priority. If the quality of the link capability of a cellular connection is predicted to be low at a given location and at a given point in time, the computer 30 may determine that the probability is adverse for failing over to the next link.

According to an embodiment, the safety re^¬ quired in a given situation depends on the nature of operation, and in this context it will also depend on whether reliable connectivity is needed to remain safe. For this, the computer 30 may determine that the predicted future link capability will directly impact the safety, such that if the link quality is estimated to be low, the computer 30 can also determine that there is a high probability that the required safety level will not be met.

The connection quality may be determined us^¬ ing a Hidden Markov Model, HMM, by the computer 30. The computer 30 may use three possible values for the quality of the connection: low, medium and high. They are based on available information that is received from the systems 100, 101, 102. The computer 30 may achieve one step ahead prediction (for example, the one step time interval may be equal to 30 minutes) .

The HMM is a type of stochastic modelling and can be used for piecewise stationary processes. The computer 30 may be set to assume stationarity of the environment by the 30 min time window. This algorithm may estimate the non-measurable variable (hidden state) applying observations from the sources. The method is widely used for example in signal processing (speech recognition) , machine learning (face recognition) and in neuroscience (DNA sequences) .

The computer 30 is configured to assume three discrete values of hidden states which correspond to the link quality: low, medium and high. The stationary environmental conditions within 30 minutes have no ef^¬ fect on the quality of the link connection. For each state the computer 30 assigns the relevant probabili^¬ ties of the state transition between two time steps, for example P (LOW | HIGH) is the probability of the transition from the current state "high" to the "low" state in the next step. It should be noticed that in the sequence of two time steps, it is possible to have the same state with relevant probability. Instead of the graph representation, the same information can be summarized in the transition matrix.

For every time step the computer 30 collects specific information associated with the state. In FIG. 6, this is depicted by reference 40 as an obser- vation received from the system 100, for example bi- trate value, signal-to-noise ratio (SNR) , etc. Proper^¬ ties of observations for each state can be described by the probability density function of receiving such an observation for the relevant state. For example in case of the good link quality (hidden state = high) , the probability of the low bit rate is close to 0, etc .

FIG. 7 illustrates an analysis of the HMM with respect to time. 30 minutes states may be used and observations Y(l...t) received for the state.

FIG. 8 summarises the calculation.

According to an embodiment, the HMM may be applied as follows.

In operation 1, historical measurements or detections are received from the connection systems. For example the computer 30 may receive historical measurements such as cellular service, SNR, latency and bandwidth, etc. as discussed in the embodiments above relating to the connection capability and safe^¬ ty.

In operation 2 declarations of historical da- ta may be applied. For example information about cel^¬ lular service disruptions or peak hour measurements such that the computer 30 has further measurements and information about effects that may have direct or in^¬ direct impact to the wireless connections.

In operation 3, in order to predict the capa^¬ bility of a given connection link in the future, the computer 30 uses previous measurements with the cur^¬ rent declarations such as warned cellular service dis^¬ ruptions or peak hours. These current declarations will then be weighted accordingly depending on relevance and impact. HMM is used in the operation 3 pre^¬ dict the results.

According to an embodiment, the computer 30 is configured by the set of instructions 38 to perform the HMM process for determining the safety and the cost characteristics and apply them for determining the route 108,109. The HMM can be described by the following set of parameters:

The list of hidden states is demonstrated by states = { LOW, MEDIUM, HIGH}. The state transition matrix is as follows

wherein all probabilities do not depend on time .

The probability of densities is as follows:

wherein all probabilities do not depend on time, and describe the distribution of observations emitted from each state.

The vector P(0) of the initial probabilities is as follows

The computer 30 is configured to update the state estimation, for example as follows

based on current observation Y(t) in time t. The computer 30 performs this further as follows

wherein the computer 30 can determine this equation recursively, for example that current probability re^¬ quires analogical values from previous step.

The computer 30 is configured to predict the future state as follows

wherein X(t+1) is the future state, and Y(l),...Y(t) ob^¬ servation from first state to t-states. This is illus^¬ trated in FIG. 7 that shows an analysis of the HMM in time. According to an embodiment, a state may be 30 minutes.

A current state is in time instant (t) and in previous step (t-1) the computer 30 has made the pre^¬ dictions that in time t the hidden state will be as follows :

wherein this probability of state LOW in time instant t is based on all previous observations

wherein this probability of state MEDIUM in time in- stant t is based on all previous observations

wherein this probability of state HIGH in time instant t is based on all previous observations (Y (1) , .... Y (t- 1) ) .

In time t the computer 30 receives new infor^¬ mation Y(t) which is use to update the predictions from previous operation. Bayesian update for state LOW is as follows

wherein this is updated probability that current hid^¬ den state is LOW based on new measurement Y(t) .

Bayesian update for state MEDIUM is as follows

wherein this is updated probability that current hid^¬ den state is MEDIUM based on new measurement Y(t) .

Bayesian update for state HIGH is as follows

wherein this is updated probability that current hid^¬ den state is HIGH based on new measurement Y(t) .

Based on these probabilities the computer 30 can make prediction for the next operation: Probabil- ity that in step (t+1) the hidden state will be "LOW" is

for above equation the computer 30 may further perform transient probabilities in order to make prediction as follows

and knowledge from previous calculations.

Probability that in step (t+1) the hidden state will be "MEDIUM" is as follows

Probability that in step (t+1) the hidden state will be "HIGH" is as follows

FIG. 8 illustrates a summary of the calcula^¬ tion process. For state (X(t)) in time "t", the com^¬ puter 30 may update the estimation from a previous state "t-1" by the information from a new observation Y(t) and make a new prediction for time instant state "t+1". The prediction is based on the determination as described above.

Alternatively, or in addition, the functionality described herein can be performed, at least in part, by one or more hardware logic components.

The term 'computer', 'computing-based de^¬ vice', 'apparatus' or 'device' is used herein to refer to any device with processing capability such that it can execute instructions. Those skilled in the art will realize that such processing capabilities are in^¬ corporated into many different devices and therefore the terms 'computer' and 'computing-based device' each include different types of computer devices, for exam^¬ ple, servers, cloud computers, or any other computing devices that are enabled for the SA system.

Although the subject matter has been described in language specific to structural features and/or acts, it is to be understood that the subject matter defined in the appended claims is not neces- sarily limited to the specific features or acts de^¬ scribed above. Rather, the specific features and acts described above are disclosed as embodiments of imple- meriting the claims and other equivalent features and acts are intended to be within the scope of the claims .

It will be understood that the benefits and advantages described above may relate to one embodi^¬ ment or may relate to several embodiments. The embod^¬ iments are not limited to those that solve any or all of the stated problems or those that have any or all of the stated benefits and advantages. It will fur- ther be understood that reference to 'an' item refers to one or more of those items.

The steps of the method described herein may be carried out in any suitable order, or simultaneous^¬ ly where appropriate. Additionally, individual blocks may be deleted from any of the methods without depart^¬ ing from the spirit and scope of the subject matter described herein. Aspects of any of the embodiments described above may be combined with aspects of any of the other embodiments described to form further embod- iments without losing the effect sought.

The term 'comprising' is used herein to mean including the method, blocks or elements identified, but that such blocks or elements do not comprise an exclusive list and a method or apparatus may contain additional blocks or elements.

It will be understood that the above descrip^¬ tion is given by way of example only and that various modifications may be made by those skilled in the art. The above specification, examples and data provide a complete description of the structure and use of exem^¬ plary embodiments. Although various embodiments have been described above with a certain degree of particu- larity, or with reference to one or more individual embodiments, those skilled in the art could make nu^¬ merous alterations to the disclosed embodiments with^¬ out departing from the spirit or scope of this speci- fication.

Claims

1. An apparatus, comprising:

a processor; and

a storage comprising a set of instructions that, when executed by the processor, cause the apparatus to:

receive first data regarding reliability of a first wireless connection;

receive second data regarding cost characteristics of a first wireless connection;

receive third data regarding reliability of a second wireless connection;

receive fourth data regarding cost characteristics of a second wireless connection;

receive a route plan of a ship configured to navigate and control the ship, wherein the first and the second wireless connections are used to control the route plan and navigate the ship;

modify the route plan based on the first, second, third and fourth data so as to process the reliability and cost characteristics of the wireless connections for the route plan, characterised by

predict the reliability and cost characteristics of the first and the second wireless connections in rela^¬ tion to time, wherein the route plan is based on the prediction.

2. The apparatus of claim 1, wherein the set of in^¬ structions, when executed by the processor, further cause the apparatus to: quantify a risk factor for the reliability and cost characteristics of the first and the second wireless connections in relation to time.

3. The apparatus of any preceding claim, wherein the reliability of the wireless connection comprises a de^¬ gree of properties with respect to delivery of data to the intended recipient; or a probability level indi- eating how reliably the wireless connections can pro^¬ vide data communications.

4. The apparatus of any preceding claim, wherein the set of instructions, when executed by the processor, further cause the apparatus to: determine a cell based network for determining the route, wherein cells indicate the reliability of the wireless connections.

5. The apparatus of any preceding claim, wherein the set of instructions, when executed by the processor, further cause the apparatus to: determine an estimate for the first data and the third data; compare the es^¬ timate to the first and the third data so as to assess the route plan.

6. The apparatus of any preceding claim, wherein the set of instructions, when executed by the processor, further cause the apparatus to: plan and schedule transmission of large amounts of data so that the transmission cost is minimized.

7. The apparatus of any preceding claim, wherein the reliability is based on one of the following charac^¬ teristics of the wireless connection: link redundancy, coverage, capability such as throughput or latency, age of measurements, signal strength, weather, moving objects with a certain height and ability to navigate around them, or stability.

8. The apparatus of any preceding claim, wherein the cost characteristics are based on one of the follow- ing: a function of an expected amount of data to be transferred, connectivity technology used, or cost of a fallback to another wireless connection.

9. The apparatus of any preceding claim, wherein the set of instructions, when executed by the processor, further cause the apparatus to: navigate the ship ac^¬ cording to the modified route plan.

10. The apparatus according to any preceding claim, wherein the first wireless connection comprises mobile communication; or wherein the second wireless connec^¬ tion comprises satellite communication.

11. The apparatus according to any preceding claim, further including a third wireless connection which comprises Wi-Fi communication;

wherein the set of instructions, when executed by the processor, cause the apparatus to:

receive fifth data regarding reliability of the third wireless connection;

receive sixth data regarding cost characteristics of the third wireless connection;

wherein the third wireless connection is further used to control the route plan and navigate the ship; and modify the route plan further based on the fifth and sixth data so as to process the reliability and cost characteristics of the wireless connections for the route plan.

12. A ship operator comprising the apparatus of any preceding claim.

13. A method, comprising:

receiving first data regarding reliability of a first wireless connection;

receiving second data regarding cost characteristics of a first wireless connection;

receiving third data regarding reliability of a second wireless connection;

receiving fourth data regarding cost characteristics of a second wireless connection;

receiving a route plan of a ship configured to navi^¬ gate and control the ship, wherein the first and the second wireless connections are used to control the route plan and navigate the ship;

modifying the route plan based on the first, second, third and fourth data so as to process the reliability and cost characteristics of the wireless connections for the route plan; and

predicting the reliability and cost characteristics of the first and the second wireless connections in rela- tion to time, wherein the route plan is based on the prediction .

14. A computer program, comprising a set of instructions which are configured to cause a computer to per- form the steps of claim 13 when executed.