Patent Application
20220281337
The present disclosure is related to a system and a method for positioning of a stationary assembly and a mobile assembly relative to each other.
For automated charging of vehicles (referred to as mobile assembly), the vehicle has to park at a position within a predefined charging range of the conductive or wireless charging device (referred to as stationary assembly). This charging range is typically smaller than 500 by 500 mm. To help the driver or the autonomous vehicle to reach this position, the charging system typically includes a positioning system with a range up to a few meters (i.e. 0.5 m-10 m), which estimates the position of the charging area with respect to the vehicle or vice versa.
The positioning system typically comprises transmitters and receivers that are either integrated in the mobile assembly, the stationary assembly, or in both assemblies. Based on the transmitted and received signals the relative and/or absolute position with respect to each other can be determined.
Often the environment around the stationary assembly can distort the signals used by the positioning system. For example, when a positioning system is used based on magnetic fields, magnetic or electric conducting materials in the vicinity can dampen or reflect the magnetic fields. In case of a system based on ultra-sound, reflection of sound from nearby objects can distort the received signals. As a result, the positioning system, calibrated in a laboratory environment, provides position information that deviates from the real position in the installed environment.
Next to the generally applied positioning system there is often other trajectory information present on the mobile assembly, such as wheel rotation, cameras for autonomous driving, etc.
It is known from WO 2014/183926, 20 Nov. 2014 to align a vehicle to a stationary power transfer unit by supplying this additional trajectory information that is gathered by vehicle borne sensors, such as rotational speed sensors, accelerometers, gyroscopes, etc. as redundant information to the park assistance control unit while executing the trajectory of the vehicle towards the stationary unit. The park assistance control unit can then correct its trajectory based on the redundant information. While this method allows for obtaining higher positioning accuracy for the instant trajectory, the calculation of a subsequent trajectory in a subsequent parking manoeuvre will not be improved.
The present disclosure aims at solving at least one and preferably all disadvantages of systems known in the art. This may be achieved by combining this additional trajectory information from the mobile assembly to improve (a posteriori) the positioning estimation of the mobile assembly with respect to the stationary assembly each time a positioning sequence is carried out. A possible application is the positioning of an electric vehicle, an automated guided vehicle, an autonomous vehicle, or a drone. Systems and methods of the present disclosure are particularly useful in relation to positioning a vehicle for conductive and/or wireless power transfer.
It is therefore an aim of the disclosure to provide a positioning system and method that is more accurate and/or efficient in determining positioning information for guidance of a mobile assembly towards a stationary assembly.
According to the present disclosure, the distortion of the position information by the environment is not removed. Rather, the (fixed) environmental distortion is taken into account by compensating the calculated position over time, e.g. iteratively. This can be done by using learning techniques such as machine learning.
The distortion compensation is improved by combining the calculated trajectory from the positioning system, referred to as first positioning system, with measurement data originated from one or more second sensors on the mobile assembly different from the first positioning system. These one or more second sensors are advantageously not used by the first positioning system in calculating the trajectory. These one or more second sensors are advantageously complementary to the sensors of the positioning system, and the second sensors are exploited to form a second positioning system that is used to calibrate and/or adapt the trajectory determined by the first positioning system. Advantageously, from the data of the one or more second sensors a second trajectory of the mobile assembly relative to the stationary assembly is determined. The combination of the measured signals of both positioning systems is used for improvement of the position estimation of the first positioning system, with the intention to learn and/or calibrate the environment of the first positioning system. As an additional advantage, after some iterations, it becomes possible to obtain absolute position data from measurements of the first positioning system when adjusted by a calibration offset obtained through multiple second trajectories determined by the so-called second positioning system.
According to a first aspect of the present disclosure, there is therefore provided a system for positioning a mobile assembly relative to a stationary assembly, as set out in the appended claims.
Systems for positioning a mobile assembly relative to a stationary assembly according to the present disclosure comprise a first position sensing device, a controller, and at least one sensor attached to the mobile assembly. The first position sensing device is configured for sensing first data representative of a relative position between the mobile assembly and the stationary assembly based on a wireless signal, which can be an electromagnetic signal (e.g. radio waves, a magnetic field, or optical signal), an ultrasound or acoustic signal, or any other suitable signal. The controller may be configured to determine the relative position or a first trajectory from a first position remote from the stationary assembly to a second position proximate to the stationary assembly based on such first data. The at least one sensor is arranged for collecting second data representative of a displacement executed by a mobile assembly. The controller is further configured for determining a calibration offset for the first trajectory based on the second data. Such calibration offset may comprise environmental distortion data (e.g. a map) configured for improving the first trajectory, such as (relative) position offset data. The calibration offset is determined following completion of the displacement of the mobile assembly from the first position (101) remote from the stationary assembly to the second position (102) proximate to the stationary assembly along the first trajectory (103). The controller may further be configured for iteratively improving the calibration offset following completion of subsequent displacements of the mobile assembly from a third to a fourth position, which fourth position preferably coincides with the second position. Optionally the controller is further configured for compensating the first trajectory based on the calibration offset.
Embodiments of such systems may further be configured for guiding the displacement of the mobile assembly from the first position to the second position, e.g. based on the compensated first trajectory. For such embodiments the controller may be configured for determining a first trajectory for guiding the mobile assembly from the first position to the second position. In such embodiment the first trajectory or the compensated first trajectory is determined before executing a displacement of the mobile assembly towards the stationary assembly. The controller typically requires information regarding the estimated relative position between the mobile and the stationary assembly and thus may be configured to determine the compensated first trajectory based on the first data and the calibration offset. The controller may further be configured to determine the first trajectory based on additional data, such as the orientation of the mobile assembly relative to the stationary assembly and/or the minimal turning radius of a mobile assembly. Such embodiments of the system may also facilitate autonomously moving the mobile assembly along the first trajectory.
In an alternative embodiment, the controller may be configured for guiding the mobile assembly from the first position to the second position based on the relative position, which may be determined based on the first data and for instance a calibration offset (e.g. the compensated first trajectory). The controller may further be configured for determining a first trajectory based on the first data. In such embodiments only the relative position is used for guiding the mobile assembly, for instance by visualizing the relative position of the stationary assembly on a display to an operator operating the mobile assembly. The mobile assembly is then displaced towards the stationary assembly for instance by the operator. Subsequently, the first trajectory is determined by the controller for instance by recording the first data during or after the displacement.
Such first data of a wireless signal may for instance comprise a parameter of the wireless signal, such as latency of receiving a wireless signal (e.g. a time of arrival of a wireless signal, a time difference of a time of arrival of wireless signals for instance originating from or arriving at multiple sensors), a signal strength, an amplitude of a signal, a phase of a signal or combinations of these. The controller may be configured to use one or more methods (e.g. multilateration, triangulation) to determine a relative position based on such first data. Some types of wireless signals may be influenced by environmental factors (e.g. temperature, pressure). Therefore, the system may comprise additional sensors for determining these environmental factors, which provide additional data to the controller for improving determining a relative position based on the first data.
Different methods may be used to determine a relative position based on the wireless signal. In general, the controller determines a relative position using a calculated position of one or more sources relative to the position of one or more sensors. In case of an ultrasound based signal, the system may be realized by providing a mobile assembly and a stationary assembly comprising one or more ultrasound transducers, wherein one of the assemblies may comprise a transducer (e.g. transmitter) transmitting an ultrasound signal and the other one of the assemblies may comprise a transducer (e.g. sensor) for receiving an ultrasound signal. Such a wireless signal has a known (or approximately known) travel speed through air. The controller may be configured to determine the time of arrival of the ultrasound signal at one or more sensors and apply multilateration to calculate the position of the transmitter(s) relative to the sensor(s). Alternatively, to eliminate the need of synchronization between transmitter and sensor, the controller may be configured to determine the time difference of arrival per sensor pair and apply multilateration to calculate the position of the transmitter(s) relative to the sensor(s). Either way, the first position sensing device may comprise additional sensors such as temperature or pressure sensors, which may provide such information regarding the environment to the controller for an improved determination of the relative position, because the speed of sound can be determined more accurately. In situations where the beam pattern of the sensor and transmitter are uniform, the relative signal strength may also be used as first data.
In embodiments where the wireless signal comprises a radio wave a similar method or controller as for ultrasound signals may be used to determine the relative position. For example, for Ultra Wide Band the first data may typically comprise a time of arrival or a time difference of arrival as first data and the controller may be configured for determining a relative position based on multilateration. Alternatively, for Wifi or Bluetooth wireless signals typically the first data will comprise a signal strength and the controller may be configured for determining a relative position based on multilateration using the first data.
Embodiments wherein the wireless signal comprises a magnetic field will typically use low frequent magnetic field (up to 150 kHz) signals. The mobile or the stationary assembly may be configured as a transmitter with one or more coils that may be oriented orthogonal to one another. The other assembly may comprise sensors for sensing the magnetic field. Typically different sensors may be used for different axes (x,y,z). These sensors may comprise simple coils (e.g. configured to measure an induced voltage), Hall effect sensors or magneto-resistance sensors to measure the magnetic field. The relative position can be determined in various ways. For example, the first data may comprise a relative signal strength and the controller may be configured for determining a relative position based on multilateration or an inverse magnetic model using the first data. The latter may be based on a theoretical Biot Savart model which expresses the magnetic field originating from a current carrying wire as a function of position in a 3D space. By inversing this model, the measured magnetic field is the input and the position in 3D space is the output. The first data may additionally comprise a phase of the wireless signal and the controller may be configured for using this additional first data in the same inverse magnetic model. Alternatively, the first data comprises a phase of the wireless signal and the controller may be configured for determining a relative position based on triangulation using the first data.
The at least one second sensor is arranged for collecting second data representative of a displacement executed by the mobile assembly and can be any sensor useful to this end. By way of example, the at least one sensor is one or more of: an accelerometer, a gyroscope, a rotation angle encoder, a rotational speed sensor and a steering angle sensor, i.e. sensors that are typically present on mobile assemblies such as vehicles and the like. The at least one second sensor advantageously does not form part of the position sensing device, and the data obtained therefrom (the second data) is advantageously not used in determining the first trajectory. The second data is therefore distinct from the first data.
According to the present disclosure, the controller is further configured for determining a calibration offset for the first trajectory based on the second data. The controller is configured for determining the calibration offset for the entire first trajectory between the first position and the second position. Advantageously, the calibration offset is determined following completion of the displacement of the mobile assembly in execution of the first trajectory from the first position to the second position, e.g. when the mobile assembly has reached or arrived at the stationary assembly and/or in the second position. By determining the calibration offset only when the mobile assembly has arrived at a final destination, it is possible to take the entire trajectory between the first position and the second position into account when compensating distortion of the first trajectory. This is not possible when a piecewise calibration of the trajectory is performed, e.g. between each two consecutive intermediate positions along the first trajectory. An improved distortion compensation is hence obtained, which may provide faster convergence.
The calibration offset comprises environmental distortion data, such as (relative) position offset data, forming a map. This map for instance comprises information regarding distortions of the first data at various locations within an area of the environment. Such calibration offset is derived from the difference between the first trajectory and the second. After multiple displacements from a first position to a second position the calibration offset comprises information regarding multiple approaches (e.g. distance, orientation) of the mobile assembly towards the stationary assembly. The map is thus updated iteratively, preferably for every approach, and can be implemented as a multi-dimensional lookup table for instance if data for translations and rotations of the mobile assembly are stored as a as a function of the first trajectory. In subsequent approaches the distortion data from this map is used to adjust/improve the first trajectory for instance based on an improved position estimation from the first data. In situations where the first trajectory is new in sense that no distortion data is present in the map for the estimated positions, interpolation of the calibration data may be used to obtain derived distortion data for the estimated positions. Subsequent displacements may improve the calibration offset either by increasing the number of estimated positions for which distortion data is collected or by updating the distortion data for estimated positions already present in the map.
Advantageously, the controller is configured for determining a second trajectory representative of the displacement that was executed by the mobile assembly, i.e. from the first position to the second position. Advantageously, the first data is not used for determining the second trajectory. The second trajectory is determined advantageously exclusively based on the second data. The controller is further configured for determining the calibration offset based on a difference between the first trajectory and the second trajectory.
In order to determine the calibration offset, the controller is advantageously configured to let the first trajectory and the second trajectory coincide in a mark position, which can be, though need not be, the second position. The mark position is advantageously a position determined by a third sensor configured to collect third data representative of the mark position. The third sensor is advantageously different from the first position sensor. As a result, the mark position will be a position comprised both in the first trajectory and the second trajectory. The second trajectory is advantageously constructed by using the mark position as common or coinciding point with the first trajectory. The second trajectory can then be calculated back from the mark position, or form the second position, e.g. when the mark position and second position are adjacent positions, to the first position, e.g. through a co-ordinate transformation. This allows for accurately reconstructing the second trajectory and for obtaining an accurate relation between the first and second trajectories useful for determining the calibration offset. The mark position is advantageously a point which is shared between the first and second trajectories, e.g. because the mobile assembly must in any case pass through the mark position to reach the second position. Alternatively, or in addition, either the first data, the second data or both comprise timestamp information related to an execution of the first trajectory or the second trajectory respectively. The timestamp information allows for relating the two trajectories in the mark position. Timestamps between the mobile assembly (second sensor) and the position sensing device can be synchronized in the first position, the second position, the mark position, and/or any other suitable position.
Advantageously, the controller is configured to determine the second trajectory after the mobile assembly has been guided/positioned in the mark position, or in the second position. The second position is advantageously a (final) park position of the mobile assembly, such as a position in registration with the stationary assembly. Such a park position is one allowing for transferring utilities between the mobile assembly and the stationary assembly. By so doing, the two trajectories share a common position, making it easy to determine the calibration offset.
The first position is advantageously a starting position for a guidance of the mobile assembly to the stationary assembly (e.g., the second position). The first position can refer to a position sensed for the first time by the first position sensor for the respective mobile assembly. The first or starting position advantageously corresponds to a most remote position of the mobile assembly that is sensed by the wireless signal of the position sensing device, or the position in which the wireless signal determines a position of the mobile assembly for the first time, e.g. following communication initialization between mobile and stationary assembly. This can be a position in which first contact is made between the mobile assembly and the stationary assembly, e.g., where wireless communication between the mobile and stationary assembly is initiated. The first position advantageously corresponds to a position where a guidance path, i.e. the first trajectory, is determined for the first time.
Advantageously, the controller is configured to collect fourth data representative of an orientation of the mobile assembly relative to the stationary assembly. The controller is configured to determine the orientation in at least one position along the displacement, e.g. in the first position, the second position, the mark position or any other suitable position. The controller can e.g. be configured to determine the second trajectory based on the fourth data. Hence, the controller can be configured to determine the calibration offset based on (e.g., taking into account) the orientation. This improves the accuracy in determining the second trajectory and may lead to faster convergence. The fourth data can be collected by a fourth sensor different from the first, second and third sensors. Alternatively, any one or a combination of the first, second or third sensors can be used for collecting the fourth data.
The controller advantageously implements a machine learning algorithm or an iterative convergence algorithm. Such algorithms are advantageously used for determining or adjusting the calibration offset in an adaptive way, e.g. based on multiple sets of first and second trajectories that have been collected. This ensures a smooth convergence of the adapted first trajectory (adapted by the calibration offset) towards the real trajectory and/or towards accurate relative or absolute positions.
Advantageously, the controller is configured for adapting a subsequent first trajectory by the calibration offset. Based on the calibration offset, the controller can be configured to determine a distortion compensation for the position estimations by the position sensing device. The distortion compensation can be used by the controller when determining subsequent trajectories.
The controller is advantageously implemented in the stationary assembly.
Advantageously, either one or both the controller part that implements determining the first trajectory and the controller part that implements determining the second trajectory are located in the stationary assembly.
The stationary assembly can refer to a ground assembly for wireless power transfer, or a docking station for wired power transfer. The mobile assembly can refer to any vehicle comprising a battery that requires charging through an external source, e.g. a battery operated vehicle, an automated guided vehicle, etc.
According to a second aspect of the present disclosure, there is provided a method for positioning a mobile assembly relative to a stationary assembly, as set out in the appended claims. Methods according to the present disclosure comprise a step of determining a first trajectory for positioning the mobile assembly at a second position proximate to the stationary assembly, e.g. from a first position remote from the stationary assembly, based on a wireless signal sensing a relative position between the mobile assembly and the stationary assembly. The wireless signal can be generated by the first position sensor as described above. The mobile assembly is displaced from the first position to the second position based on the first trajectory. Following completion of the displacement, a calibration offset for the first trajectory is determined. The calibration offset is determined based on sensor data other than the wireless signal and collected on or within the mobile assembly during execution of the displacement. The sensor data advantageously refers to the second data collected by the one or more second sensors as described above.
Advantageously, a second trajectory corresponding to the displacement that was executed, in particular from the first position to the second position, is determined based on the sensor data. Advantageously, the calibration offset is determined based on a difference or deviation between the first trajectory and the second trajectory.
Advantageously, the method can comprise registering a mark position of the mobile assembly relative to the stationary assembly, e.g. by collecting third data different from the wireless signal, wherein the mark position is determined based on the third data. In determining the calibration offset, the first trajectory and the second trajectory are made to coincide in the mark position. The mark position can be, though need not be, the second position. Advantageously, timestamp information is collected corresponding to at least a portion, or all of: the first data, the second data, or both. The timestamp information facilitates making the first and second trajectories coincide in the mark position.
Advantageously, methods according to aspects of the present disclosure comprise collecting fourth data representative of an orientation of the mobile assembly relative to the stationary assembly. The second trajectory is advantageously determined taking the fourth data into account. The calibration offset can hence be determined based on the fourth data. The fourth data can be different from any one or all of the wireless signal, the second data, the third data.
Advantageously, the calibration offset is determined by feeding the first trajectory and the second trajectory to a machine learning algorithm or an iterative convergence algorithm. By way of example, for a first trajectory Ti a respective calibration offset COi is determined. The calibration offset COi is used for correcting/adapting a subsequent first trajectory Ti+1.
Advantageously, methods according to the present disclosure reflect implementations of the controller of the system described above. Alternatively, the system as described above is made to execute the method steps as described herein.
1. System (30) for positioning a mobile assembly (10) relative to a stationary assembly (20) or system for guiding a mobile assembly towards a stationary assembly, comprising:
wherein the controller (31) is configured for determining a calibration offset (105) for the first trajectory (103) based on the second data, the controller being configured for determining a calibration offset following completion of the displacement of the mobile assembly from the first position (101) to the second position (102) along the first trajectory (103) and wherein the controller is configured for guiding the mobile assembly (10) from a first position (101) remote from the stationary assembly to a second position (102) proximate to the stationary assembly based on the first data and the calibration offset.
2. System of clause 1, wherein the controller is configured to determine a second trajectory (104) representative of the displacement executed by the mobile assembly based on the second data, the controller being configured for determining the calibration offset (105) based on a difference between the first trajectory (103) and the second trajectory (104).
3. System of clause 2, comprising means for registering a mark position of the mobile assembly relative to the stationary assembly along the displacement, wherein the controller (31) is configured to determine the calibration offset by making the first trajectory (103) and the second trajectory (104) coincide in the mark position.
4. System of clause 3, wherein the means for registering the mark position comprises a third sensor configured to collect third data representative of the mark position, the third sensor being different from the position sensing device.
5. System of clause 3 or 4, wherein the second position is the mark position.
6. System of any one of the preceding clauses, wherein the first position sensing device (40) is configured to emit an electromagnetic signal or an ultrasound signal for sensing the first data.
7. System of any one of the preceding clauses, wherein the second position (102) is a parking position of the mobile assembly (10).
8. System of clause 7, wherein the second position allows for transferring utilities between the mobile assembly and the stationary assembly (20).
9. System of clause 8, wherein the transfer of utilities comprises a transfer of electrical energy.
10. System of any one of the preceding clauses, wherein the controller (31) implements a machine learning algorithm or an iterative convergence algorithm for determining a distortion compensation for the first position sensing device (40) from the calibration offset (105).
11. System of any one of the preceding clauses, wherein the controller (31) is configured for adapting a subsequent first trajectory based on the calibration offset (105).
12. System of any one of the preceding clauses, wherein the at least one sensor (51, 53) is one or more of: an accelerometer, a gyroscope, a rotation angle sensor, a rotational speed sensor and a steering angle sensor.
13. System of any one of the preceding clauses, wherein the system is configured to collect fourth data representative of an orientation of the mobile assembly relative to the stationary assembly, wherein the controller is configured to determine the orientation in at least one position along the displacement, and to determine the calibration offset based on the orientation.
14. System of any one of the preceding clauses, wherein sensing first data based on the wireless signal comprises measuring a parameter of the wireless signal representative of the relative position.
15. Stationary assembly (20) comprising the system of any one of the preceding clauses or for use in a system of any one of the preceding clauses.
16. Method (200) for positioning a mobile assembly (10) relative to a stationary assembly (20) comprising the steps of:
17. Method of clause 16, comprising determining a second trajectory (104) representative of the displacement executed by the mobile assembly based on the sensor data, wherein determining the calibration offset comprises evaluating a difference between the first trajectory (103) and the second trajectory (104).
18. Method of clause 17, comprising registering a mark position of the mobile assembly relative to the stationary assembly along the displacement, wherein determining the calibration offset (105) comprises making the second trajectory coincide with the first trajectory (103) in the mark position.
19. Method of clause 18, comprising collecting third data different from the wireless signal, and using the third data for registering the mark position.
20. Method of clause 18 or 19, wherein the mark position is the second position.
21. Method of any one of the clauses 16 to 20, wherein the second position (102) is a parking position of the mobile assembly.
22. Method of clause 21, wherein the parking position allows for transferring utilities between the mobile assembly and the stationary assembly.
23. Method of clause 22, wherein transferring utilities comprises transferring electrical energy.
24. Method of any one of the clauses 16 to 23, comprising collecting fourth data representative of an orientation of the mobile assembly relative to the stationary assembly, and determining the calibration offset based on the fourth data.
25. Method of clause 24 in conjunction with any one of the clauses 17 to 20, comprising determining the second trajectory based on the fourth data.
26. Method of any one of the clauses 16 to 25, wherein a distortion compensation is determined by feeding the calibration offset (105) to a machine learning algorithm or an iterative convergence algorithm.
27. Method of any one of the clauses 16 to 26, comprising adjusting the first trajectory (103) based on the calibration offset (105) when determining a subsequent first trajectory.
28. Method of any one of claims 15 to 27, wherein sensing the relative position comprises measuring a parameter of the wireless signal.
Aspects of the present disclosure will now be described in more detail with reference to the appended drawings, wherein same reference numerals illustrate same features and wherein:
Referring to
Referring to
In one example, the control unit 31 is provided on the stationary assembly 20, although this is not a requirement and it may well be possible to provide the control unit 31 on the mobile assembly 10. The trajectory 103 determined by the control unit 31 is transmitted wirelessly to the mobile assembly 10, e.g. through a wireless communication interface 33 including communicating antennas 32, 52, through which the trajectory 103 is communicated to a control unit 50 on the mobile assembly. Control unit 50 may be configured to supply driving or guidance instructions to the mobile assembly 10, or to an operator driving the mobile assembly, based on the trajectory 103. It will be convenient to note that, alternatively, the trajectory 103 can be calculated in control unit 50 based on position data fed from the position sensors (e.g. receivers 42) to the control unit 50.
The signal from the position sensors of position sensing system 40 can be distorted by the environment. For example, when a positioning sensing system is used based on magnetic fields, magnetic or electric conducting materials in the vicinity can dampen or reflect the magnetic fields. In case of a system based on ultra-sound, reflection of sound from nearby objects can distort the received signals. As a result, the positioning sensing system, calibrated in a laboratory environment, provides position information that deviates from the real position in the installed environment.
According to the present disclosure, the accuracy of the position calculation is improved by calibration of the environment and compensation of the calculated position based on this calibration. The present disclosure uses sensor data of a secondary system provided on the mobile assembly 10, which describes (part of) the executed relative trajectory, to calibrate or adjust the trajectory 103 determined by the control unit 31.
To this end, the mobile assembly 10 is equipped with sensors 51, 53, such as accelerometers, (rotational) speed sensors, gyroscopes, wheel encoders or other sensing systems that are used for park assist, park distance control or autonomous driving, etc., which are typically present on mobile assemblies, such as vehicles. Sensors 51, 53 are coupled to the control unit 50 which may process data received from these sensors to determine an executed trajectory 104 which may deviate from the calculated trajectory 103 (see
The sensors 51, 53 are advantageously used in the present disclosure as a secondary position sensing system to determine a second trajectory 104 relating to the path that is actually executed by the mobile assembly 10. This trajectory 104 data is then used to correct or calibrate the calculated trajectory 103 following completion of the displacement of the mobile assembly to the final park position 102.
In one aspect, since the second trajectory 104 is determined based on sensors located on the mobile assembly 10, a relative trajectory is obtained. To determine an error between the actually executed trajectory 104 and the calculated trajectory 103, both trajectories are advantageously brought in registration in a mark position 106. The mark position advantageously refers to a position where the positioning system 30 can obtain an accurate position information which is not affected by the environment. The mark position 106 can e.g. be determined by a beacon 60 which does not form part of the position sensing system 40 and/or the sensors 51, 53. The mark position 106 can be selected to be a location where the distance between the mobile assembly 10 and the stationary assembly 20 is very small, e.g. when the mobile assembly 10 is in registration with the stationary assembly 20, e.g. the mark position 106 can coincide with, or be adjacent to parking position 102 though this need not be the case, as e.g. shown in
Advantageously, an orientation a of the mobile assembly 10 is determined along the executed path. The orientation is determined in at least one position along the first trajectory 103, the second trajectory 104 or both, and can be used to improve determination of the respective trajectory relative to the stationary assembly 20. The orientation a can be determined through data gathered by any one of the position sensing system 40, the sensors 51, 53 or any other (additional) sensor.
Control unit 50 is advantageously configured to communicate data representative of the executed trajectory 104 only after the mobile assembly 10 has arrived at final position 102. The positioning system 30 can derive a calibration offset 105 based on the calculated/estimated trajectory 103 and the executed trajectory 104, as determined by data collected by sensors 51, 53. Because the stationary assembly 20 is typically fixed at a position in a garage or on a driveway, the calibration offset 105 can be improved over time when the mobile assembly 10 parks multiple times at position 102, and/or from calibration offsets determined from multiple mobile assemblies. This will incrementally take (fixed) environment influences into account. Preferably, the control unit 31 can be implemented with an iterative convergence algorithm or a machine learning algorithm that is used to gradually adjust and improve the calculation of trajectory 103 based on the calibration offset 105. A gradual convergence algorithm is preferred in order to restrict the influence of outliers in the position estimation of both the first as well as the second positioning system. In other words, this prevents large deviations between subsequent position estimations.
Referring to
In step 202, the first trajectory 103 is determined. Control unit 31 can activate the position sensing system 40 to determine the position 101 at which the mobile assembly is positioned. From that position, control unit 31 determines a first trajectory 103 which the mobile assembly 10 should follow to arrive at final parking position 102. Trajectory 103 is communicated to control unit 50 on the mobile assembly 10. Control unit can additionally determine an orientation of the mobile assembly 10 in position 101, e.g. by means of position sensing system 40 and use the orientation in determining trajectory 103.
In step 203, the mobile assembly 10 is displaced from position 101 to position 102. The control unit 50 can guide the mobile assembly 10, either through autonomous driving or operated-assisted driving to execute the first trajectory 103. The position sensing system 40 may continuously guide the mobile assembly, by updating the position of the mobile assembly 10 and adapting the first trajectory 103 accordingly. While executing the first trajectory 103, control unit 50 collects data relating to the actual executed trajectory, e.g. from sensors 51, 53.
In step 204, a mark position 106 is determined along the actual executed path, e.g. by beacon 60. By way of example, a timestamp at which the mobile assembly 10 passes through mark position 106 is determined, and mark position 106 can refer to a known position relative to the stationary assembly 20 (second position 102). Additionally, or alternatively, an orientation a of the mobile assembly 10 is determined along the path. Mark position 106 and/or orientation a is fed to control unit 31.
In step 205, the mobile assembly 10 reaches the final parking position 102 and the first trajectory 103 is completed. Control unit 50 may communicate the collected trajectory data received from sensors 51, 53 to the control unit 31. Alternatively, the data from sensors 51, 53 may be communicated to the control unit 31 during execution of the first trajectory 103.
In step 206, control unit 31 determines a calibration offset 105. Either control unit 50, or control unit 31 can determine a second trajectory 104 which is representative of the path actually executed by the mobile assembly 10. The data collected from sensors 51, 53, and optionally the orientation a is used for determining the second trajectory 104. Second trajectory 104 includes position 102, e.g. as end point of the trajectory and can include mark position 106 which may or may not overlap with position 102. Advantageously, no data from position sensing system 40 is used to determine the second trajectory 104. Control unit 31 can compare the second trajectory 104 to the first trajectory 103 of the position sensing system 40 to derive the calibration offset.
A difference between the first trajectory 103 and the second trajectory 104 can be determined once the mobile assembly has arrived at the final parking position 102. The first trajectory 103 and the second trajectory 104 are advantageously made to coincide in the mark position 106, and a deviation between the two can easily be determined. By way of example, the first trajectory 103 and the second trajectory 104 are brought in registration in the mark position 106. As the time at which the mobile assembly 10 reached the mark position 106 is known, the entire second trajectory 104 can be related to the first trajectory 103. Depending on the accuracy of the position estimation of the first trajectory, the positioning estimation of the position sensing system 40 can be updated (re-calibrated) using machine learning techniques, or iterative convergence algorithms, with the purpose of improving subsequent positioning estimations and/or trajectory determination of the position sensing system 40.
| Number | Date | Country | Kind |
|---|---|---|---|
| 2023068 | May 2019 | NL | national |
| Filing Document | Filing Date | Country | Kind |
|---|---|---|---|
| PCT/EP2020/062351 | 5/4/2020 | WO |
