JP2025524594A - Attitude readjustment method for dead reckoning systems using relative positioning systems - Google Patents

Abstract

The present invention relates to an attitude readjustment method (107) including a step (116) of estimating an estimated position of the dead reckoning system (22) in an arbitrary fixed coordinate system at a plurality of determination times by the dead reckoning system (22), a step (114) of obtaining an estimated position of the dead reckoning system (22) in a predetermined fixed coordinate system at each determination time as estimated by a relative positioning system (24), and a step (124) of deriving at least one attitude readjustment parameter by minimizing a cost function that compares the estimated position with an estimated position corrected by each readjustment parameter.
[Selected figure] Figure 5

Description

The present invention relates to dead reckoning technology, and in particular to attitude readjustment technology provided by dead reckoning systems. The present invention has advantageous application in urban or indoor environments, i.e., when traveling within buildings.

It is now common to track the location of an object using multilateral positioning, measuring the distance between a receiver attached to the object and at least three reference points with known positions in the environment. This can take advantage of positioning systems such as GNSS (Global Navigation Satellite System, e.g., GPS) or the infrastructure of wireless communication networks (e.g., Wi-Fi networks, GSM networks, etc.). However, these methods are very limited because the availability and accuracy of the information cannot be guaranteed due to masking that can occur between the reference points and the receiver. As a result, their use in urban or indoor environments requires the deployment of costly infrastructure with numerous reference points distributed throughout the environment. Furthermore, they rely on external technologies such as GNSS satellites, which may be unavailable or intentionally jammed.

Alternatively, methods known as dead reckoning (DRR) are known that track the relative position of an object in any environment using motion sensors that measure the object's movement. Relative position refers to the object's position in space relative to a point and coordinate system given at initialization. In addition to position, these methods can obtain the object's orientation (also known as "attitude") relative to the same initial coordinate system, given by Euler angles (roll φ, pitch θ, yaw ψ) in three dimensions, or heading ψ in two dimensions. These methods are suitable for movement in environments where position tracking using multilateral positioning is difficult, such as urban areas and indoor environments.

There are many different types of dead reckoning. The most common is what is called "simple" inertial navigation, implemented in heavy applications such as fighter jets, passenger aircraft, submarines, and ship navigation. It is generally based on an inertial unit consisting of at least three accelerometers and three gyrometers arranged in three axes. In general, the gyrometers "maintain" a frame of reference, and motion can be estimated by double time integrating the accelerometer measurements in this frame. It is well known that very high-precision sensors are required to use this "simple" inertial navigation. In fact, the double time integration of acceleration measurements means that a constant acceleration error produces a position error that increases proportional to the square of the time.

Another dead reckoning technique is to provide velocity vector information in the object's coordinate system from an external source (e.g., a car's odometer, a boat's log, or an airplane's pitot tube). By applying a simple integral to this velocity vector information and combining it with the object's attitude, particularly its heading, its trajectory can be determined. For the same sensor measurement error, the effect of time drift is reduced.

In most cases, the initial attitude is known, for example, by the initial "alignment" of the inertial system or by sensors other than inertial sensors (e.g., magnetic sensors). However, measurement errors in the inertial sensors cause time drift in the measured attitude, which, depending on the accuracy of the sensors used, can invalidate the initial attitude information over time, resulting in an inaccurate object position. For example, a 1% error in heading measurement can result in a 1m error in the object's position after 100m of movement.

To address this, it is known to periodically recalibrate the measurements of a dead reckoning system with a separate positioning system in order to best maintain the object's attitude information, and in particular its orientation information. For example, from WO 2019/020961 it is known to recalibrate the orientation information using magnetic orientation measurements obtained by a magnetometer mounted on the object.

However, this solution is not entirely satisfactory. In fact, magnetometers have their own errors and drift issues, so their measurements may not be reliable enough to use as a basis for recalibration. This solution also requires that magnetometers be integrated into the object whose position is to be tracked, which increases costs.

One object of the present invention is to propose a simple and economical solution for readjusting the attitude provided by a dead reckoning system. Another object is to enable such readjustment to be performed with high accuracy over short distances. Another object is to enable accurate tracking of people moving around inside a building in a simple and economical way.

To this end, the present invention provides, according to a first aspect, an attitude readjustment method for readjusting an attitude provided by a dead reckoning system, comprising:
a step of estimating, by the dead reckoning system, an estimated position of the dead reckoning system in an arbitrary fixed coordinate system at the determined time for each of a plurality of determined times included in a movement section in which the dead reckoning system moves along the movement trajectory;
for each determination time instant, obtaining an estimated position of the dead reckoning system in a predetermined fixed coordinate system at the determination time instant, as estimated by the relative positioning system;
and deriving at least one attitude readjustment parameter by minimizing a cost function that compares the evaluated position with an estimated position corrected by each readjustment parameter.

According to certain embodiments of the present invention, the posture readjustment method has one or more of the following features, taken alone or in any technically possible combination:

The dead reckoning system is worn by the pedestrian, preferably on the pedestrian's foot or ankle.

The dead reckoning system comprises a motion sensor for measuring the movement of the dead reckoning system and a processing unit for deriving the attitude and position of the dead reckoning system from the measured movement.

The relative positioning system is selected from a multi-angle positioning system, a multi-plane positioning system, a map matching system, and a visual positioning system, and preferably, the relative positioning system includes an ultra-wideband telemetry device.

The arbitrary fixed coordinate system and the predetermined fixed coordinate system have a common axis, and the attitude readjustment parameters are composed of parameters, preferably angles, for correcting the attitude by rotation around the common axis.

The common axis is the vertical axis.

The cost function represents the average geometric deviation between the evaluated and estimated positions after applying geometric transformations, including rotation and preferably translation, to the evaluated or estimated positions.

Rotation occurs around an axis of rotation, and translation occurs in a direction perpendicular to the axis of rotation.

The step of deriving the readjustment parameter includes calculating a candidate value for the readjustment parameter, evaluating a precision value for the candidate value, comparing the precision value with a previous precision value associated with the previous readjustment parameter, and determining the readjustment parameter in response to the result of the comparison, where the readjustment parameter depends on the candidate value and the previous readjustment parameter.

The accuracy value is a function of the uncertainty of the estimated and evaluated positions and/or the average geometric deviation between the evaluated and estimated positions after applying candidate values of the realignment parameters.

The step of deriving the retuning parameters includes:
a) calculating first candidate values for the realignment parameters by minimizing a cost function comparing evaluated positions for the N decision times with estimated positions for the N decision times corrected by the realignment parameters;
b) calculating a first accuracy value associated with the first candidate value;
c) calculating a second candidate value for the realignment parameter by minimizing a cost function that compares the evaluated positions for N-1 decision times (N decision times minus the oldest decision time) with the estimated positions for the N-1 decision times corrected by the realignment parameter;
d) calculating a second accuracy value associated with the second candidate value;
e) comparing the first accuracy value with the second accuracy value;
f) selecting the first candidate value if the first accuracy value reflects the best accuracy.

If the accuracy value reflecting the best accuracy is constituted by the second accuracy value, the step of deriving the retuning parameters comprises:
a sub-step of deleting the evaluation position and the estimated position for the oldest decision time;
and the sub-step of decrementing N by 1 and repeating the sub-steps a) to e).

Dead reckoning systems have an accuracy of a few percent, e.g., 1-3%, for distance traveled, and an accuracy of tens of degrees per hour, e.g., 30-80 degrees per hour, for azimuth drift, while relative positioning systems have an accuracy of tens of centimeters, e.g., 20 cm to 1 m, for position.

The present invention also provides, according to a second aspect, a method for determining the position of an object carrying a dead reckoning system in a predefined space, the method comprising:
activating a dead reckoning system;
recalibrating the attitude of the dead reckoning system using a relative positioning system including infrastructure installed at access points to a predefined space to obtain attitude recalibration parameters, by implementing any of the attitude recalibration methods described above;
recalibrating the dead reckoning system using a relative positioning system including infrastructure installed at access points to a predefined space to obtain repositioning parameters;
calculating, by the dead reckoning system, a calculated position of the dead reckoning system in a predetermined coordinate system using the attitude readjustment parameters and the position readjustment parameters.

According to a particular embodiment of the present invention, the location determination method also has the following features:

The infrastructure is mounted on the vehicle itself, equipped with a position and attitude system capable of calculating the infrastructure's position in a given coordinate system.

According to a third aspect, the present invention relates to a dead reckoning system comprising a motion sensor for measuring the movement of the dead reckoning system and a processing unit for deriving the attitude and position of the dead reckoning system in a fixed coordinate system from the measured movement, the processing unit being configured to execute the attitude readjustment method according to the first aspect in order to readjust the attitude.

According to a fourth aspect, the present invention relates to a computer program product comprising code instructions for performing the attitude readjustment method according to the first aspect when the program is executed by a processor.

Finally, according to a fifth aspect, the present invention relates to storage means readable by a computer device having stored thereon a computer program product including code instructions for executing the attitude readjustment method according to the first aspect.

Other features and advantages of the present invention will become apparent from the following description, given by way of example only, and taken in conjunction with the accompanying drawings.

1 is a top view of a system for locating an object within a predefined space in accordance with an illustrative embodiment of the present invention; FIG. 2 is a perspective view showing details of the location system of FIG. 1; FIG. 2 illustrates a location box of the location system of FIG. 1. 2 illustrates an example of a method for locating an object within a predefined space implemented by the system of FIG. 1; FIG. 5 illustrates a reorientation step of the method of FIG. 4 . FIG. 6 illustrates substeps of deriving reorientation parameters of the reorientation step of FIG. 5 .

The location system 10 shown in FIG. 1 is for locating an object 12 within a predefined space 14. To this end, the location system 10 includes a plurality of location boxes 16, each attached to an object 12. The location system 10 also includes a location infrastructure 18 located at at least one access point 19 to the predefined space 14.

Referring to FIG. 2, each object 12 is here a pedestrian. The present invention is particularly advantageous in such applications, as the box 16 takes up little space and can be easily and ergonomically worn by the pedestrian. Alternatively (not shown), the object 12 can be any moving object for which position information is required, such as, for example, a wheeled vehicle or a drone.

The predefined space 14 is typically the interior of a building. For example, in the case of an industrial site, the pedestrians 12 are engineers working at the site. Alternatively, if the predefined space 14 is an intervention site, such as a fire site or a hostage situation, the pedestrians 12 are firefighters or infantrymen intervening at the site.

Each localization box 16 is typically worn on a limb of the walker 12, here a leg, preferably a foot or ankle. For this purpose, each localization box 16 is provided with a harness 20, for example a wristband with a hook-and-loop fastener, which encircles the limb and allows a fixed connection, as can be seen in FIG. 3. Alternatively (not shown), the harness 20 can be constituted by any element capable of fixedly connecting the localization box 16 to the object 12.

Referring further to FIG. 3, the location box 12 includes a dead reckoning system 22 and a relative positioning system 24. In the illustrated example, the location box 12 also includes a communication system 26, typically a wireless communication system, for communication with an external device, such as a mobile terminal 29 (FIG. 2), e.g., a multifunction mobile terminal, or a remote server (not shown). It also optionally includes a storage module 28.

The dead reckoning system 22 comprises a motion sensor 30 for measuring the movement of the dead reckoning system 22, and a processing unit 32 for deriving from the measured movement the attitude and position of the dead reckoning system 22 in a predetermined fixed coordinate system, for example the East-North-Up (ENU) coordinate system. "Fixed coordinate system" means a coordinate system that is fixed in the ground reference system.

In the example described here, the dead reckoning system 22 is configured to operate in a two-dimensional coordinate system and is configured to provide only the orientation of the dead reckoning system 22 in the horizontal plane of the given coordinate system and two position coordinates. Alternatively (not shown), the dead reckoning system 22 may be configured to operate in a three-dimensional coordinate system and is configured to provide the roll, pitch, and yaw angles of the dead reckoning system 22 in the given coordinate system and its three position coordinates. Those skilled in the art will be able to easily translate the examples shown for the two-dimensional case into the three-dimensional case.

The motion sensor 30 includes a gyrometer 40 for measuring the angular velocity of the dead reckoning system 22 according to a system having three orthogonal axes defining a moving coordinate system fixed to the box 16, i.e., for measuring the three components of an angular velocity vector in this moving coordinate system. It will therefore be understood that the gyrometer 40 may in fact be a set of three gyrometers associated with one of the three axes, in particular a three-axis gyrometer (i.e., each capable of measuring one of the three components of the angular velocity vector).

The motion sensor 30 also includes an acquisition unit 42 for acquiring the linear velocity of the dead reckoning system 22, i.e., the linear velocity of its movement. This acquisition unit 42 can acquire the linear velocity directly or indirectly, and therefore various types can be used.

For example, the acquisition unit 42 can consist of one or more accelerometers (not shown). These accelerometers are advantageously arranged in three axes according to the same system with three orthogonal axes as the gyrometer 40. They are sensitive to external forces other than gravity applied to the sensor 30 and can measure a specific acceleration. The linear velocity is determined by integrating this acceleration with time.

Alternatively, if the object 12 is a wheeled vehicle, the acquisition unit 42 can consist of at least two odometers corresponding to the wheels of the vehicle, for example the two rear wheels. An odometer is a device ("rev counter") that can measure the wheel speed by counting the number of wheel revolutions. Typically, an odometer has a component (e.g., a magnet) fixed to the wheel, and counts the number of revolutions per unit time (rotation frequency) by detecting each passing of this component (called the "top"). Other techniques are known, such as optical detection of marks on the wheel or magnetometers, such as those described in French Patent Application Publication No. 2939514, that detect the rotation of a metal object such as a wheel. Here, the "speed" of the wheel is a scalar, i.e., the norm of the wheel's speed in the ground reference frame (assuming no skid). If the radius of the wheel is known, the speed norm can be estimated by measuring the rotation frequency.

Optionally, the motion sensor 30 may include a stride detector (not shown) for detecting when the feet of the walker 12 touch the ground, as described, for example, in WO 2017/060660.

In the illustrated example, the processing unit 32 is constituted by a programmable machine such as a DSP (Digital Signal Processor) or a microcontroller. It comprises a processor or CPU (Central Processing Unit) 44 and a memory 46, such as a RAM (Random Access Memory) and/or a ROM (Read Only Memory). The processor 44 is configured to execute instructions loaded into the memory 46. When the dead reckoning system 22 is powered on, the processor 44 can read and execute instructions from the memory 46. These instructions form a computer program that causes the processor 44 to calculate the orientation and position of the dead reckoning system 22 in a predetermined fixed coordinate system, for example by implementing the methods described in WO 2017/060660 and FR 2 939 514.

In a variant (not shown), the processing unit 32 can be formed by a dedicated machine or component, such as an FPGA (Field-Programmable Gate Array) or an ASIC (Application-Specific Integrated Circuit).

The processing unit 32 also includes a buffer memory 49 for temporarily storing information necessary to calculate the orientation and position of the dead reckoning system 22 in a predetermined fixed coordinate system.

Optionally, the dead reckoning system 22 comprises a network of spatially spaced magnetometers 48 linked to the box 16, i.e., having a motion substantially identical to the motion of the box 16 in the ground reference frame. Each magnetometer 48 is a three-axis magnetometer capable of measuring magnetic fields along three axes. To this end, each magnetometer 48 typically comprises three single-axis magnetometers (not shown) oriented along axes that are substantially perpendicular to one another. These axes are preferably the same as the axes of the three orthogonal axes of the gyrometer 40.

Due to its special geometry, the network of magnetometers 48 is able to determine, at each measurement time of the magnetometers 48, the spatial gradient of the measured magnetic field, in particular the coefficients of this gradient along each axis of the moving coordinate system linked to box 16. The gradient coefficients can be determined, for example, by a method using the vector measurements of the magnetic field by the magnetometers 48 in combination with an optimization method of the least squares or median filter type, or in combination with the intrinsic properties of the magnetic field described by Maxwell's equations. However, other conventional methods adapted to calculate the coefficients of the spatial gradient of the magnetic field can also be used.

The processing unit 32 is also typically configured to adjust the velocity determination of the dead reckoning system 22 by implementing the method described in EP 2 541 199 A1.

The dead reckoning system 22 typically has an accuracy of a few percent, e.g., 1-3%, for distance traveled, and an accuracy of tens of degrees per hour, e.g., 30-80 degrees per hour, for heading drift.

The relative positioning system 24 is capable of determining the relative position of a point fixed to the dead reckoning system 22 with respect to a reference system whose position in a predetermined fixed coordinate system is known, and of deriving therefrom the position of the dead reckoning system 22 in the predetermined fixed coordinate system. For this purpose, the relative positioning system 24 comprises a sensor 50 suitable for measuring parameters that enable the positioning of the fixed point with respect to the reference system, and a processing unit 52 for deriving from these parameters the position of the dead reckoning system 22 in the predetermined fixed coordinate system.

Preferably, the relative positioning system 24 is constituted by a multilateral positioning system, in particular a proximity multilateral positioning system. Furthermore, the infrastructure 18 comprises at least three beacons 54 (FIG. 2) (or "anchors") at the access points 19, each having a known position in a predetermined fixed coordinate system, with which the relative positioning system 24 is configured to communicate. To this end, the sensor 50 is typically constituted by a wireless communication system capable of communicating with the beacons 54 and estimating the distance to each beacon 54, for example by measuring two-way ranging (TWR) distances.

Preferably, the sensor 50 comprises an ultra-wideband telemetry device capable of communicating with the beacon 54 and measuring the distance to the beacon 54 via an ultra-wideband (UWB) protocol. This allows for position measurement accuracy on the order of tens of centimeters, e.g., 20 cm to 1 m. As known to those skilled in the art, the UWB protocol is a wireless communication protocol based on the transmission of very short pulses (on the order of nanoseconds) across a wide frequency spectrum. This allows for communication over a wide bandwidth (500-1,350 MHz) with a center frequency of 0.5-9.5 GHz, depending on the channel used. Alternatively, the sensor 50 can communicate with the beacon 54 via a Bluetooth protocol or a Wi-Fi protocol.

As a variant, the multilateral positioning system can be constituted by a GNSS system.

Furthermore, the processing unit 52 is configured to derive the relative position of the sensor 50 in a predetermined fixed coordinate system from the distance measurements and the known position of the beacon 54, typically by multilateral positioning or optimization, and to derive the position of the dead reckoning system 22 in the predetermined fixed coordinate system from this position and the position of the sensor 50 relative to the dead reckoning system 22.

According to another embodiment (not shown), the relative positioning system 24 is constituted by a multi-angle positioning system. The sensor 50 is capable of measuring at least one angle between the observation directions of two beacons, each of which has a known position in a predetermined fixed coordinate system. The processing unit 52 is further configured to derive the position of the sensor 50 in the predetermined fixed coordinate system from the angle measurements and the known positions of the beacons 54, typically by multi-angle positioning, and to derive the position of the dead reckoning system 22 in the predetermined fixed coordinate system from this position and the position of the sensor 50 relative to the dead reckoning system 22.

According to yet another embodiment (not shown), the relative positioning system 24 is implemented as a map-matching system. The sensor 50 can measure environmental parameters, such as terrain or magnetic fields, and the processing unit 52 can match these measurements with a map of parameters stored in memory, thereby deriving the position of the sensor 50 in a predetermined fixed coordinate system.

According to a fourth embodiment (not shown), the relative positioning system 24 is constituted by a visual positioning system. The sensor 50 is constituted by an imager associated with an image processing system. The imager is configured to acquire images of the environment, and the processing system is configured to detect, in each image, feature points, such as target markers, whose positions are known in a predetermined fixed coordinate system. The processing unit 52 is further configured to derive the relative position of the sensor with respect to the feature points, and to derive the position of the dead reckoning system 22 in the predetermined fixed coordinate system from this relative position, the known positions of the feature points, and the position of the sensor 50 relative to the dead reckoning system 22.

In a variant (not shown), the imager may be fixed (mounted on infrastructure 18) and feature points (typically target markers) may be attached to box 16. In a further variant, multiple targets may be attached to box 16, providing a motion capture configuration that is detected by the fixed imager.

In the illustrated example, the processing unit 52 is implemented by a programmable machine such as a DSP (Digital Signal Processor) or a microcontroller. It comprises a processor or CPU (Central Processing Unit) 56 and memory 58, such as RAM (Random Access Memory) and/or ROM (Read Only Memory). The processor 56 is configured to execute instructions loaded into the memory 58. When the relative positioning system 24 is powered on, the processor 56 can read and execute instructions from the memory 58. These instructions form a computer program that causes the processor 56 to calculate the orientation and position of the dead reckoning system 22 in a predetermined fixed coordinate system from parameters measured by the sensors 50.

In a variant (not shown), the processing unit 52 can be formed by a dedicated machine or component, such as an FPGA (Field-Programmable Gate Array) or an ASIC (Application-Specific Integrated Circuit).

In the illustrated example, the processing unit 52 of the relative positioning system 24 is separate from the processing unit 32 of the dead reckoning system 22. Alternatively (not shown), these processing units 32, 52 may be integrated.

The communication system 26 is configured to perform short-range wireless communication, e.g., Bluetooth or Wi-Fi, and/or to connect to a mobile network (e.g., UMTS/LTE/5G) for long-range communication (particularly with the mobile terminal 29 in one embodiment). Alternatively (not shown), the communication system 26 may be a wired connection technology (e.g., USB) for transferring data from the storage module 28 to another storage module, e.g., the mobile terminal 29.

For example, the communication system 26 is configured to transmit the position calculated by the dead reckoning system 22 to the mobile device 29, which then displays the position on the interface of its navigation software.

In the above example, the processing units 32, 52 of the dead reckoning system 22 and the relative positioning system 24 are integrated in the box 16. As a variant (not shown), at least some of these processing units 32, 52 may be located, for example, in the mobile terminal 29, in the infrastructure 18, and/or in a remote server (not shown). In other words, at least some of the steps of calculating the position of the dead reckoning system 22 by the dead reckoning system 22 or the relative positioning system 24 may be performed by the mobile terminal 29, the infrastructure 18, and/or the remote server. Furthermore, the communication system 26 may be configured to transmit data from the motion sensor 30 and/or data from the sensor 50 to the mobile terminal 29, the infrastructure 18, and/or the remote server. Advantageously, the communication system 26 is further configured to receive the position of the dead reckoning system 22 calculated by the dead reckoning system 22 or the relative positioning system 24 from the mobile terminal 29, the infrastructure 18, and/or the remote server.

Returning to Figures 1 and 2, the infrastructure 18, as described above, comprises a plurality of beacons 54 positioned at access points 19 to the predefined space 14. These beacons 54 are typically integrated into portals positioned at the access points 19, such as the portal 60 shown in Figure 2. Each portal 60 comprises at least three beacons 54 so that pedestrians 12 passing through the access points 19 can be located by multilateral positioning.

The infrastructure 18 may be, for example, a permanent, fixed infrastructure. This is particularly true when the predefined space 14 is an industrial site and the localization system 10 is intended to track the movements of technicians within the site. Alternatively, the infrastructure 18 may be a temporary, fixed infrastructure. This is true, for example, when the predefined space 14 is an intervention site, in particular a fire site, and the infrastructure 18 is transported and installed at one of the intervention site's access points before the firefighters arrive. Alternatively, the infrastructure 18 may be a mobile infrastructure. This is typically mounted on a vehicle (not shown) used to transport the pedestrians 12, and the pedestrians 12's movements to the predefined space 14 are tracked. This vehicle itself is equipped with a position and orientation system capable of calculating the position of the infrastructure 18 in a predetermined coordinate system. This is true, for example, when the predefined space 14 is an intervention site, in particular in the event of a hostage situation.

The method 100 implemented by the location system 10, and more particularly the processing units 32 and 52, will now be described with reference to Figures 4 to 6.

As shown in FIG. 4, the method 100 begins with a first step 102 of activating the dead-reckoning system 22. This first step 102 is typically performed while the pedestrian 12 wearing the box 16 is still outside the predefined space 14. Typically, step 102 is initiated by the pedestrian 12 pressing a button (not shown) on the box 16. The relative positioning system 24 is typically activated simultaneously with step 102.

The dead reckoning system 22 also provides the pedestrian 12's orientation ψ(t) in the aforementioned arbitrary fixed coordinate system. This arbitrary fixed coordinate system is a horizontal two-dimensional coordinate system, and is typically constructed by a rotation of an angle θ around the vertical axis of the predetermined fixed coordinate system and a translation of a horizontal vector Δ. Therefore, the arbitrary fixed coordinate system shares the vertical axis as a common axis with the predetermined fixed coordinate system. This is selected arbitrarily by the dead reckoning system 22 depending on the attitude at the time of startup.

Step 102 is followed by step 104, which determines whether the relative positioning system 24 can estimate the position of the dead-reckoning system 22 in a predetermined fixed coordinate system. If the determination is positive, i.e., typically if the box 16 is within range of a beacon 54 of the infrastructure 18, step 104 is followed by step 106, which readjusts the position of the dead-reckoning system 22, and step 107, which readjusts the orientation of the dead-reckoning system 22. If the determination is negative, i.e., typically if the box 16 is out of range of a beacon 54 of the infrastructure 18, step 104 is repeated after a period of time has elapsed.

Referring to FIG. 5, the orientation readjustment 107 begins with the pedestrian 12 moving 110 along the movement trajectory during a movement segment while the box 16 is within range of a beacon 54 of the infrastructure 18.

During the movement 110, in the orientation readjustment 107, for a plurality of evaluation times i _k included in the movement section, the relative positioning system 24 estimates 112 the position u(i _k ) of the pedestrian 12 in a predetermined fixed coordinate system at each evaluation time i _k . After the estimation 112, the dead reckoning system 22 receives 114 the estimated position u(i _k ).

In parallel, in the orientation readjustment 107, for a plurality of estimated times τ _k included in the movement section, the dead reckoning system 22 estimates 116 the position v(τ _k ) of the pedestrian 12 in an arbitrary fixed coordinate system at each estimated time τ _k .

After the receiving step 114 and the estimating step 116, a step 118 is performed in which the estimated position u( _ik ) and the estimated position v( _tk ) by the dead reckoning system 22 are matched. The readjusting step 107 is based on the use of estimated positions synchronized with the estimated positions. However, these two types of positions come from different sources, and the estimated time _ik is generally different from the estimated time _tk . Therefore, matching of the estimated positions u( _ik ) and the estimated positions v( _tk ) is necessary. This matching aims to associate and synchronize the estimated positions u( _ik ) and v( _tk ) so that, for multiple determination times _{tk ,} there is a pair of estimated positions u( _tk ) and estimated positions v( _tk ) at each determination time tk. Therefore, this matching is performed by finding the following from a set {ui _k } _k of evaluation positions u(i _k ) at evaluation time i _k and a set {v(τ _k )} _k of estimated positions v(τ _k ) at estimation time τ _k :
A set {u(t _k )} _k of evaluation positions u(t _k ) at decision time t _k , and
It consists in deriving a set {v(t _k )} _k of estimated positions v(t _k ) at decision time t _k .

Therefore, matching can be done, for example,
A set {u( _ik )} _k of evaluation positions u( _ik ) at evaluation time i _k , and
Interpolating at least one set or subset of the set {v(τ _k )} _k of estimated positions v(τ _k ) at estimated times τ _k .

This interpolation preferably uses time splines. The splines used for interpolation are preferably at least twice differentiable. In particular, the spline representation makes it possible to enforce continuity of the trajectory. It also makes it possible to optimize the various relevant parameters, in particular the time synchronization parameters, using methods based on the gradient of the criterion to be minimized. Alternatively, the interpolation can be performed using other approaches, such as, for example, a discrete representation of the trajectory traveled by the pedestrian 12.

Preferably, only the set of estimated positions {v( _τk )} _k is interpolated, and the decision time _tk is selected to be equal to the evaluation time _ik ( _tk = _ik ). Thus, the set of evaluated positions u( _tk ) {u( _tk )} _k at the decision time _tk is merged with the set of evaluated positions u( _ik ) {u( _ik )} _k at the evaluation time _ik . In this case, receiving 114 the evaluated position u( _ik ) at the evaluation time _ik by the dead reckoning system 22 constitutes the step of obtaining the evaluated position u( _tk ) of the pedestrian 12 at the decision time _tk by the dead reckoning system 22.

Alternatively, only the set of evaluation positions {u(t _k )} _k may be interpolated, and the decision time t _k may be chosen to be equal to the estimated time τ _k (t _k =τ _k ). In this case, matching 118 comprises obtaining, by the dead reckoning system 22, the estimated position u(t _k ) of the pedestrian 12 at the decision time t _k .

Note that if evaluation time i _k corresponds to evaluation time τ _k , then matching 118 is limited to associating the evaluation position u(i _k ) and the estimated position v(τ _k ) with the same time i _k , τ _k , which becomes the decision time t _k .

After matching 118, the matched pairs of positions u( _tk ), v( _tk ) are added 120 to a buffer memory 49, and the buffer memory 49 is cleaned 122. In the cleaning 122, pairs of positions u( _tk ), v( _tk ) associated with a determination time _tk that is not within the movement interval are deleted from the buffer memory 49. The movement interval is understood here as a sliding time window of a predetermined duration that ends on the date of the most recent evaluation time _ik or estimation time _τk .

After steps 120 and 122, step 124 is performed to derive an orientation readjustment parameter θ. Here, the orientation readjustment parameter θ is constituted by a parameter, in particular an angle, of the orientation correction due to rotation around an axis of a predetermined coordinate system (in this example, the vertical axis). Furthermore, orientation readjustment 107 may include, in parallel with step 124, another step (not shown) to derive another orientation readjustment parameter, for example a parameter for correcting the deviation of the angular change of the orbit.

6, step 124 includes a first sub-step 130 of calculating a first candidate value θ1 for the orientation readjustment parameter θ using all points, i.e., all pairs of positions u( _tk ), v( _tk ), contained in buffer memory 49. This first candidate value _θ1 is calculated by minimizing a cost function that compares evaluated positions u( _tk ) at N determination times _tk corresponding to pairs of positions u( _tk ), v( _tk ) contained in buffer memory 49 with estimated positions v( _tk ) at said N determination times _tk corrected by the orientation readjustment parameter θ and _the position readjustment parameter Δ. This cost function represents, in particular, the average geometric deviation between the evaluated positions { _uk } _k and the estimated positions { _vk } _k after applying the following geometric transformation to the evaluated positions { _uk } _k :
a rotation about an axis of a given coordinate system (here the vertical axis) by an angle equal to the reorientation parameter θ, and a vector translation equal to the reorientation parameter Δ in a direction perpendicular to said axis.

where {u _k } _k is the set of evaluation positions at the decision time.
{v _k } _k is the set of estimated positions at the decision time.
{t _k } _k is the set of decision times.
N is the number of pairs of positions u(t _k ), v(t _k ) contained in the buffer memory 49 .
θ is the reorientation parameter.
R(θ) is a rotation matrix that represents the attitude modification due to application of the reorientation parameters.
Δ is the repositioning parameter.

Therefore, the first candidate value θ ₁ is as follows:
where u ₁ (t _k ) is a first position coordinate along a first axis of a predetermined fixed coordinate system of the estimated position of the dead reckoning system 22 at determination time t _k .
u ₂ (t _k ) is a second position coordinate along a second axis of the predetermined fixed coordinate system of the estimated position of the dead reckoning system 22 at determination time t _k .
v ₁ (t _k ) is a first position coordinate along a first axis of an arbitrary fixed coordinate system of the estimated position of the dead reckoning system 22 at determination time t _k .
v ₂ (t _k ) is a second position coordinate along a second axis of an arbitrary fixed coordinate system of the estimated position of the dead reckoning system 22 at determination time t _k .
N is the number of pairs of positions u(t _k ), v(t _k ) contained in the buffer memory 49 .

This sub-step 130 is followed by a sub-step 132 of evaluating a first accuracy value σ _θ1 associated with the first candidate value θ _1. The first accuracy value σ _θ1 is, for example, a function of the uncertainties σ _v , σ _u of the estimated position {v _k } _k and the evaluated position {u _k } _k . It typically consists of the standard deviation of the first candidate value θ ₁ and is given by the following formula:
where {u _k } _k is the set of evaluation positions at the decision time.
{v _k } _k is the set of estimated positions at the decision time.
{t _k } _k is the set of decision times.
σ _u ² is the variance of the random error affecting each coordinate of the position estimated by the relative positioning system 24 .
σ _v ² is the covariance of the random error affecting each coordinate of the position estimated by the dead reckoning system 22 .
u ₁ (t _k ) is a first position coordinate along a first axis of a predetermined fixed coordinate system of the estimated position of the dead reckoning system 22 at determination time t _k .
u ₂ (t _k ) is a second position coordinate along a second axis of the predetermined fixed coordinate system of the estimated position of the dead reckoning system 22 at determination time t _k .
v ₁ (t _k ) is a first position coordinate along a first axis of an arbitrary fixed coordinate system of the estimated position of the dead reckoning system 22 at determination time t _k .
v ₂ (t _k ) is a second position coordinate along a second axis of an arbitrary fixed coordinate system of the estimated position of the dead reckoning system 22 at determination time t _k .
N is the number of pairs of positions u(t _k ), v(t _k ) contained in the buffer memory 49 .

Alternatively, the first accuracy value σ _θ1 may be a function of the mean deviation observed between the set of evaluation positions {u _k } _k and the associated set of estimated positions {v _k } _k corrected by an orientation realignment parameter θ having a first candidate value θ ₁ , as given for example by the following equation:

where N is the number of pairs of positions u(t _k ), v(t _k ) contained in the buffer memory 49 .
{u _k } _k is the set of evaluation positions at the decision time.
{v _k } _k is the set of estimated positions at the decision time.
θ ₁ is the reorientation parameter assigned to the first candidate value θ ₁ .
R(θ ₁ ) is a rotation matrix that represents the pose modification due to application of the reorientation parameters.

Alternatively, the first accuracy value σ _θ1 may be a combination of functions of the uncertainties σ _v , σ _u of the estimated positions { _{v k} } _k and the evaluated positions {u _k } _k , and a function of the mean deviation observed between the set of evaluated positions {u k } _k and the associated set of estimated positions {v _k } _k corrected by an orientation readjustment parameter θ having a first candidate value θ ₁ .

The derivation step 124 further comprises, in parallel with _the sub-steps 130 and 132, a sub- _step 134 of calculating a second candidate value _θ2 of the orientation readjustment parameter θ without using the oldest point contained in the buffer memory 49, i.e. using all pairs of positions u( _tk ), v( _tk ) contained in the buffer memory 49 except for the pair of positions u(tk), v(tk ₎ associated with the oldest determination time t1. This second candidate value _θ2 is calculated by minimizing a cost function comparing evaluated positions u( _tk ) for the N-1 determination times _tk corresponding to the most recent pairs of positions u( _tk ), v( _tk ) contained in the buffer memory 49 with estimated positions v( _tk ) for said N-1 determination times _tk corrected by the orientation readjustment parameter θ and the position readjustment parameter Δ. This cost function represents, in particular, the average geometric deviation between the evaluation position {u _k } _k and the estimated position {v _k } _k after applying the following geometric transformation to the evaluation position {u _k } _k :
a rotation about an axis of a given coordinate system (here the vertical axis) by an angle equal to the reorientation parameter θ, and
A vector translation equal to the repositioning parameter Δ in a direction perpendicular to the above axes.

where {u _k } _k is the set of evaluation positions at the decision time.
{v _k } _k is the set of estimated positions at the decision time.
{t _k } _k is the set of decision times.
N is the number of pairs of positions u(t _k ), v(t _k ) contained in the buffer memory 49 .
θ is the reorientation parameter.
R(θ) is a rotation matrix that represents the attitude modification due to application of the reorientation parameters.
Δ is the repositioning parameter.

Therefore, the second candidate value θ ₂ is as follows:
where u ₁ (t _k ) is a first position coordinate along a first axis of a predetermined fixed coordinate system of the estimated position of the dead reckoning system 22 at determination time t _k .
u ₂ (t _k ) is a second position coordinate along a second axis of the predetermined fixed coordinate system of the estimated position of the dead reckoning system 22 at determination time t _k .
v ₁ (t _k ) is a first position coordinate along a first axis of an arbitrary fixed coordinate system of the estimated position of the dead reckoning system 22 at determination time t _k .
v ₂ (t _k ) is a second position coordinate along a second axis of an arbitrary fixed coordinate system of the estimated position of the dead reckoning system 22 at determination time t _k .
N is the number of pairs of positions u(t _k ), v(t _k ) contained in the buffer memory 49 .

This sub-step 134 is followed by a sub-step 136 of evaluating a second accuracy value σ _θ2 associated with the second candidate value θ _2. The second accuracy value σ _θ2 is, for example, a function of the uncertainties σ _v , σ _u of the estimated position {v _k } _k and the evaluated position {u _k } _k . It typically consists of the standard deviation of the second candidate value θ ₂ and is given by the following formula:
where {u _k } _k is the set of evaluation positions at the decision time.
{v _k } _k is the set of estimated positions at the decision time.
{t _k } _k is the set of decision times.
σ _u ² is the covariance of the random error affecting each coordinate of the estimated position by the relative positioning system 24 .
σ _v ² is the covariance of the random error affecting each coordinate of the position estimated by the dead reckoning system 22 .
u ₁ (t _k ) is a first position coordinate along a first axis of a predetermined fixed coordinate system of the estimated position of the dead reckoning system 22 at determination time t _k .
u ₂ (t _k ) is a second position coordinate along a second axis of the predetermined fixed coordinate system of the estimated position of the dead reckoning system 22 at determination time t _k .
v ₁ (t _k ) is a first position coordinate along a first axis of an arbitrary fixed coordinate system of the estimated position of the dead reckoning system 22 at determination time t _k .
v ₂ (t _k ) is a second position coordinate along a second axis of an arbitrary fixed coordinate system of the estimated position of the dead reckoning system 22 at determination time t _k .
N is the number of pairs of positions u(t _k ), v(t _k ) contained in the buffer memory 49 .

Alternatively, the second accuracy value σ _θ2 may be a function of the mean deviation observed between the set of evaluation positions {u _k } _k and the associated set of estimated positions {v _k } _k corrected by an orientation reorientation parameter θ having a second candidate value θ ₂ , as given for example by the following equation:

where N is the number of pairs of positions u(t _k ), v(t _k ) contained in the buffer memory 49 .
{u _k } _k is the set of evaluation positions at the decision time.
{v _k } _k is the set of estimated positions at the decision time.
θ ₂ is the reorientation parameter assigned to the second candidate value θ ₂ .
R(θ ₂ ) is a rotation matrix that represents the pose modification due to application of the reorientation parameters.

Alternatively, the second accuracy value σ _θ2 may be a combination of functions of the uncertainties σ _v , σ _u of the estimated positions { _{v k} } _k and the evaluated positions {u _k } _k , and a function of the mean deviation observed between the set of evaluated positions {u k } _k and the associated set of estimated positions {v _k } _k corrected by an orientation readjustment parameter θ having a first candidate value θ ₂ .

Substeps 132 and 136 are followed by substep 140 of comparing the first precision value σ _θ1 with the second precision value σ _θ2 .

If the first accuracy value σ _θ1 reflects better accuracy than the second value σ _θ2 , i.e., σ _θ1 < σ _θ2 , step 140 is followed by step 142 of selecting the first candidate value _θ1 as the selected candidate value.

This maximizes the accuracy of the readjustment parameter θ.

Substep 142 is optionally followed by a series of substeps 150-154 aimed at validating the new retuned parameter θ from the residual calculations.

Sub-step 142 is followed by sub-step 150 of calculating an expected realignment residual r ^exp . This realignment residual is intended to reflect the expected average deviation between the set of evaluation positions {u _k } _k and the associated set of estimated positions {v _k } _k corrected by the orientation realignment parameter θ. This is given by the following equation:
where N is the number of pairs of positions u(t _k ), v(t _k ) contained in the buffer memory 49 .
σ _u ² is the covariance of the random error affecting each coordinate of the estimated position by the relative positioning system 24 .
σ _v ² is the covariance of the random error affecting each coordinate of the position estimated by the dead reckoning system 22 .
σ _θ ² is the covariance of the retuning parameter θ, and is equal to the square of the above-mentioned standard deviation σ θ ₁ of the selected candidate value θ ₁ .
u ₁ (t _k ) is a first position coordinate along a first axis of a predetermined fixed coordinate system of the estimated position of the dead reckoning system 22 at determination time t _k .
u ₂ (t _k ) is a second position coordinate along a second axis of the predetermined fixed coordinate system of the estimated position of the dead reckoning system 22 at determination time t _k .
v ₁ (t _k ) is a first position coordinate along a first axis of an arbitrary fixed coordinate system of the estimated position of the dead reckoning system 22 at determination time t _k .
v ₂ (t _k ) is a second position coordinate along a second axis of an arbitrary fixed coordinate system of the estimated position of the dead reckoning system 22 at determination time t _k .

Sub-step 150 is followed by sub-step 152 of calculating the observed realignment residual r ^obs . This realignment residual is intended to reflect the average deviation observed between the set of evaluation positions {u _k } _k and the associated set of estimated positions {v _k } _k corrected by the orientation realignment parameter θ assigned to the selected candidate value θ _1. This is given by the following equation:
where N is the number of pairs of positions u(t _k ), v(t _k ) contained in the buffer memory 49 .
{u _k } _k is the set of evaluation positions at the decision time.
{v _k } _k is the set of estimated positions at the decision time.
θ is the reorientation parameter.
R(θ) is a rotation matrix that represents the attitude modification due to application of the reorientation parameters.

Substep 152 is followed by substep 154 of comparing the observed reconditioned residual r ^obs with the expected reconditioned residual r ^exp increased by a predetermined threshold δ.

If the observed rescaled residual r ^obs is strictly greater than the expected rescaled residual r ^exp plus the threshold δ, i.e., if the inequality r ^obs >r ^exp + δ holds, then sub-step 154 is followed by sub-step 156 of rejecting the selected candidate value. The rescaled parameter θ then retains its previous value.

If the observed rescaled residual r ^obs is less than or equal to the sum of the expected rescaled residual r ^exp and the threshold δ, i.e., if the inequality r ^obs ≦r ^exp + δ holds, sub-step 154 is followed by sub-step 158 of comparing the accuracy value σ _θn associated with the selected candidate value with the accuracy value σ _θo associated with the current value of the rescaled parameter θ. Note that, given the candidate value selection algorithm, the accuracy value σ _θn associated with the selected candidate value is equal to the first accuracy value σ _θ1 mentioned above.

More specifically, in substep 158, the accuracy value σ _θn associated with the selected candidate value is compared with the accuracy value σ _θo associated with the current value of the readjustment parameter θ increased by a time drift value proportional to the elapsed time between the determination of the selected candidate value and the determination of the current value of the orientation readjustment parameter θ, this time drift value typically being equal to (t _n −t _o )×b.
where t _n is a time representing the decision time {t _k } _k that is taken into account in determining the selected candidate value.
t ₀ is a time representing the decision time {t _k } _k taken into account in determining the current value of the orientation readjustment parameter θ.
b is a predefined constant representing a typical bias value of the gyrometer 40 .

Each time t _o , t _n may be, for example, the average of the decision times {t _k } _k considered for determining the selected candidate value and for determining the current value of the reorientation parameter θ, respectively. Alternatively, each time t _o , t _n may be, for example, the most recent decision time t _k considered for determining the selected candidate value and for determining the current value of the reorientation parameter θ, respectively.

If sub-step 158 is being performed for the first time since the dead-reckoning system 22 has been powered up, the accuracy value σ _θo associated with the current value of the realignment parameter θ is preferably infinity.

Substep 158 is followed by substep 160, which determines a future value of the readjustment parameter θ depending on the result of comparison 158. This future value is typically a function of the candidate value and the current value.

For example, if the precision value σ _θn associated with the selected candidate value is greater than or equal to the precision value σ _θo associated with the current value increased by the time drift value, i.e., if the inequality σ _θn ≥ σ _θo + (t _n - _{t o} ) × b holds, then the future value is equal to the current value. On the other hand, if the precision value σ _θn associated with the selected candidate value is strictly less than the precision value σ _θo associated with the current value increased by the time drift value, i.e., if the inequality σ _θn < σ _θo + (t _n - _{t o} ) × b holds, then the future value is equal to the selected candidate value.

Alternatively, the future value may be equal to a combination of the selected candidate value and the current value as a function of the ratio of an accuracy value σ _θn associated with the selected candidate value to an accuracy value σ _θo associated with the current value increased by the time drift, typically with a Kalman filter or Bayesian fusion applied. For example, the future value may be equal to:

where θ _n is the selected candidate value.
σ _θn is the precision value associated with the selected candidate value above.
θ _o is the current value of the reorientation parameter θ.
σ _θo is the precision value associated with the current value above.

This sub-step 160 completes step 124 of deriving the orientation readjustment parameter θ.

5, step 124 is followed by an adjusted estimated heading step 126, in which the dead reckoning system 22 adjusts the estimated heading by applying a heading readjustment parameter θ.

This step 126 completes the orientation readjustment step 107.

4, after the repositioning step 106 and the reorienting step 107, a step 109 occurs in which the dead reckoning system 22 calculates the position of the pedestrian 12. In this step 109, the dead reckoning system 22 applies the reorienting parameters and the repositioning parameters to determine the position of the pedestrian 12 in a predetermined coordinate system, which is given by the following equation:

v(t) is the position of the pedestrian 12 in an arbitrary coordinate system estimated by the dead reckoning system 22 .
θ is the reorientation parameter.
R(θ) ^T is a rotation matrix that represents the attitude modification due to application of the inverse of the reorientation parameters.
Δ is the repositioning parameter.

In parallel, method 100 returns to step 104, allowing the readjustment parameters for orientation θ and position Δ to be continuously updated each time pedestrian 12 passes near beacon 54.

The exemplary embodiment described above allows for accurate tracking of people's movements within a building in a simple and economical way. This objective is achieved by means of particularly lightweight infrastructure 18 that only needs to be deployed, and particularly simple location boxes 16 that only need to be worn by the people to be tracked. The motion sensors 30 provided on these boxes 16 do not need to be very accurate, as the dead reckoning system can be periodically recalibrated each time a person passes close to the infrastructure 18.

In particular, the application of UWB technology to the relative positioning system 24 is particularly advantageous because the small error of UWB positioning allows for very accurate recalibration over short movement ranges. This significantly reduces infrastructure requirements. Furthermore, the recalibration is completely transparent to the pedestrian 12, as they do not need to take any special steps to ensure the recalibration of their location information box 16.

Claims (15)

An attitude readjustment method (107) for readjusting attitude provided by a dead reckoning system (22), comprising:
a step (116) of estimating, by the dead reckoning navigation system (22), for each of a plurality of determined times included in a movement section in which the dead reckoning navigation system (22) moves along a movement trajectory, an estimated position of the dead reckoning navigation system (22) in an arbitrary fixed coordinate system at the determined time;
for each determined time instant, obtaining (114) an estimated position of the dead reckoning system (22) in a predetermined fixed coordinate system at said determined time instant, as estimated by a relative positioning system (24);
and deriving (124) at least one attitude readjustment parameter by minimizing a cost function that compares the evaluated position with the estimated position corrected by each readjustment parameter. The attitude readjustment method (107) described in claim 1, wherein the dead reckoning system (22) is worn by the pedestrian (12), and preferably the dead reckoning system (22) is worn on the foot or ankle of the pedestrian (12). An attitude readjustment method (107) according to claim 1 or 2, wherein the dead reckoning system (22) comprises a motion sensor (30) for measuring the movement of the dead reckoning system (22) and a processing unit (32) for deriving the attitude and position of the dead reckoning system (22) from the measured movement. An attitude readjustment method (107) according to any one of claims 1 to 3, wherein the relative positioning system (24) is selected from a multi-angle positioning system, a multi-plane positioning system, a map matching system, and a visual positioning system, and preferably the relative positioning system (24) includes an ultra-wideband telemetry device. An attitude readjustment method (107) according to any one of claims 1 to 4, wherein the arbitrary fixed coordinate system and the predetermined fixed coordinate system have a common axis, and the attitude readjustment parameter is a parameter, preferably an angle, for correcting the attitude by rotation around the common axis. The attitude readjustment method (107) described in any one of claims 1 to 5, wherein the cost function represents the average geometric deviation between the evaluation position and the estimated position after applying a geometric transformation, including rotation and preferably translation, to the evaluation position or the estimated position. The attitude readjustment method (107) of any one of claims 1 to 6, wherein the step of deriving the readjustment parameter includes calculating (130) a candidate value for the readjustment parameter, evaluating (132) an accuracy value for the candidate value, comparing (158) the accuracy value with a previous accuracy value associated with a previous readjustment parameter, and determining (160) the readjustment parameter depending on the result of the comparison (158), wherein the readjustment parameter depends on the candidate value and the previous readjustment parameter. An attitude readjustment method (107) as described in claim 7, wherein the accuracy value is a function of the uncertainty of the estimated position and the evaluated position, and/or a function of the average geometric deviation between the evaluated position and the estimated position after applying the candidate values of the readjustment parameters. The step of deriving (124) the retuning parameters comprises:
a) calculating (130) a first candidate value for the realignment parameter by minimizing a cost function comparing the evaluated positions for N decision times with the estimated positions for the N decision times corrected by the realignment parameter;
b) calculating (132) a first accuracy value associated with said first candidate value;
c) calculating (134) a second candidate value for the readjustment parameter by minimizing a cost function that compares the evaluated positions for N-1 decision times (the N decision times minus the oldest decision time) with the estimated positions for the N-1 decision times corrected by the readjustment parameter;
d) calculating (136) a second accuracy value associated with said second candidate value;
e) comparing the first accuracy value with the second accuracy value (140);
f) selecting (142) said first candidate accuracy value if said first candidate accuracy value reflects the best accuracy. If the accuracy value reflecting the best accuracy is constituted by the second accuracy value, the step of deriving (124) the retuning parameters comprises:
a substep (144) of deleting the evaluated position and the estimated position for the earliest determination time;
and a sub-step of decreasing N by 1 and repeating sub-steps a) to e). A method (100) for locating an object (12) carrying a dead reckoning system (22) in a predefined space (14), comprising:
activating (102) the dead reckoning system (22);
- a step (107) of readjusting the attitude of the dead reckoning system (22) using a relative positioning system (24) comprising an infrastructure (18) installed at access points (19) to the predefined space (14) to obtain attitude readjustment parameters, the step (107) being performed by implementing an attitude readjustment method according to any one of claims 1 to 10;
repositioning (106) the dead reckoning system (22) using a relative positioning system (24) including infrastructure (18) installed at access points (19) to the predefined space (14) to obtain repositioning parameters;
and calculating (109), by the dead reckoning system (22) using the attitude readjustment parameters and the position readjustment parameters, a calculated position of the dead reckoning system (22) in a predetermined coordinate system. The location method (100) described in claim 11, wherein the infrastructure (18) is mounted on a vehicle equipped with a position and attitude system that enables calculation of the position of the infrastructure (18) in the predetermined coordinate system. A dead reckoning system (22), comprising:
A dead reckoning system (22) comprising: a motion sensor (30) for measuring the movement of the dead reckoning system (22); and a processing unit (32) for deriving the attitude and position of the dead reckoning system (22) in a fixed coordinate system from the measured movement, wherein the processing unit is configured to execute an attitude readjustment method (107) according to any one of claims 1 to 10 in order to readjust the attitude. A computer program product comprising code instructions for performing the attitude readjustment method (107) described in any one of claims 1 to 10 when the program is executed by a processor. Storage means readable by a computer device, on which a computer program product including code instructions for executing the attitude readjustment method (107) described in any one of claims 1 to 10 is recorded.