Method of determining an alignment angle of a radar sensor for a radar auto alignment controller of a road vehicle

Abstract

A method of determining the alignment angle of two or more road vehicle (1) borne radar sensors (4) for a road vehicle radar auto alignment controller (3) starting from an initial available coarse estimate of the alignment angle. Signals related to the distance of detection, the azimuth angle and the rate of change of distance are obtained from at least two radar sensors (4). The detections are screened (5) to determine detections from stationary targets. A linearized signal processing model is derived from the determined detections from stationary targets, the model involving the alignment angle, the longitudinal and lateral velocity and the side slip rate of the road vehicle. A filtering algorithm is applied (6) to estimate the alignment angle. Based on the estimated alignment angle, a signal is generated suitable for causing the road vehicle radar auto alignment controller to perform radar offset compensation.

Description

TECHNICAL FIELD

[0001] The present disclosure relates generally to the determination of alignment angles of road vehicle radar sensors, and in particular to a method of determining alignment angles of two or more road vehicle-mounted radar sensors for a road vehicle radar auto-alignment controller, starting from an initially available coarse estimate of the alignment angles. BACKGROUND

[0002] Many road vehicles today include object detection sensors, which are used, for example, to enable localization of the ego vehicle and to enable collision warning or avoidance and other active safety applications. Such object detection sensors can use any of a variety of detection technologies, such as, for example, short or long range radar, cameras with image processing, laser or LIDAR, and ultrasound. The object detection sensors detect vehicles and other objects in the vicinity of the ego vehicle, and the application software of the localization or safety functionality can use such object detection information to take action or issue warnings when appropriate.

[0003] For such localization or safety functionality to perform optimally, the object detection sensors must be properly aligned with the ego vehicle. Misalignment of the sensors can have important consequences, as vehicles and other objects in the vicinity of the ego vehicle can be interpreted as having a different location than they actually have. Even if there are multiple object detection sensors on the vehicle, it is important to align them so as to minimize or eliminate conflicting or inaccurate sensor readings.

[0004] Previously, object detection sensors were typically integrated directly into the front or rear fascia of the road vehicle. This type of installation has the disadvantage that there is typically no practical way to physically adjust the alignment of such object detection sensors. Thus, if an object detection sensor becomes misaligned from the true heading of the vehicle, for example due to damage to the fascia or deformation related to use and weather, there is traditionally no way to correct the misalignment other than to replace the entire fascia assembly containing the sensor.

[0005] US 2017261599 (Al) presents an attempt to address this challenge and discloses a method and sensor system for automatically determining object sensor position and alignment on a host vehicle and for automatically calibrating sensor position and alignment in software, thereby ensuring accurate sensor readings without the need for mechanical adjustment of the sensors. In normal operation, a radar sensor detects objects around the host vehicle. Static objects are objects identified as having a ground speed approximately equal to zero. Vehicle dynamics sensors provide vehicle longitudinal and lateral speed and side yaw-rate data. The measured data for static objects, including azimuth angle, range, and range rate relative to the sensor, are used in conjunction with the vehicle dynamics data for recursive geometric calculations that converge on actual values for the two-dimensional position and azimuth alignment angle of the radar sensor on the host vehicle. Only static objects are used in the pose estimation calculations. By using only static objects, the number of unknowns involved in the sensor pose estimation calculations is reduced, enabling the pose of the radar sensor to be determined through recursive calculations over multiple measurement cycles. Static objects have a position defined by range and azimuth angle and range rate, all of which are measured by the radar sensor. From the defined geometric relationships, a pair of calculations can be performed recursively upon the arrival of each new set of sensor measurement data. In the first calculation, the azimuth orientation angle a is assumed to be known (from a default setting, or from a previous cycle of recursive calculations), and the position values a and b are calculated. In the second calculation, the position values a and b are assumed to be known (from a default setting, or from a previous cycle of recursive calculations), and the azimuth orientation angle a is calculated. Over a period of time (notably, a minute or several minutes), with measurement data arriving several times per second, these calculations will converge to produce actual values for the sensor pose (a, b, a).

[0006] While US 2017261599 (Al) discusses estimation of sensor mounting position and azimuth alignment angle, this is done under the assumption that longitudinal speed, lateral speed, and yaw rate are known, which enables the problem to be solved by treating the minimization problem as a constrained least squares problem. However, in many cases, longitudinal speed, lateral speed, and yaw rate are unknown, which is the reason this solution cannot be applied. Therefore, there is a need for an improved method of determining alignment angle for road vehicle carried object sensors, in particular, road vehicle carried radar sensors. SUMMARY

[0007] It is an object of the present invention to provide an improved method of determining alignment angle for two or more road vehicle carried radar sensors.

[0008] According to a first aspect, there is provided a method of determining an alignment angle of two or more road vehicle carried radar sensors for a road vehicle radar auto-alignment controller, starting from an initially available coarse estimate of the alignment angle, the method comprising: obtaining signals related to detected range, azimuth angle and range rate from at least two radar sensors; screening the detections to determine detections from stationary targets; deriving a linearized signal processing model from the determined detections from stationary targets, the model involving the alignment angle, longitudinal and lateral velocity and side slip rate of the road vehicle; applying a filtering algorithm to estimate the alignment angle; generating a signal suitable for causing a road vehicle radar auto-alignment controller to perform radar offset compensation based on the estimated alignment angle. By allowing the use of a linearized signal processing model it is enabled to apply a computationally efficient filtering algorithm to estimate the alignment angle in an efficient and rapid manner.

[0009] In another embodiment, the method further comprises obtaining the initially available coarse estimate of the alignment angle from a known nominal installation angle of the radar sensors carried by the vehicle.

[0010] In still another embodiment, the method further comprises screening the detections to determine detections from stationary targets by monitoring the longitudinal and lateral velocity and side slip rate of the road vehicle together with the obtained range rate of the detections.

[0011] In still another embodiment, the method further comprises screening the detections to determine detections from stationary targets using a so-called RANSAC technique or by running target tracking and looking at temporal effects.

[0012] In an additional embodiment, the method further comprises deriving the linearized signal processing model by performing a Taylor series expansion of a rotation matrix involving the azimuth angle alignment angle, longitudinal and lateral velocity and side slip rate of the road vehicle.

[0013] In still an additional embodiment, the method further comprises that the filtering algorithm applied to estimate the alignment angle is one of a Kalman filter, a least mean square filter, a recursive least square filter, a windowed least square filter or a filter based on other signal processing algorithms suitable for use with the linearized signal processing model.

[0014] According to a second aspect, there is a road vehicle system comprising a controller for radar auto-alignment according to the above method.

[0015] According to another aspect, there is a system comprising a controller for radar auto-alignment according to the above method.

[0016] According to still another aspect, there is a computer program embodied on a non-transitory computer readable storage medium, the computer program comprising program code for controlling a road vehicle radar auto-alignment controller to perform a process for road vehicle radar auto-alignment, the process comprising the above method.

[0017] According to still another aspect, there is a computer program comprising instructions which, when executed by processing circuitry, is configured to cause the road vehicle system to perform the above method.

[0018] According to an additional embodiment, there is a carrier comprising the above computer program, wherein the carrier is one of an electronic signal, optical signal, radio signal, or computer readable storage medium.

[0019] The above embodiments have the advantageous effect of allowing the use of a linearized signal processing model, and thereby enabling the application of computationally efficient filtering algorithms to estimate the alignment angle on a road vehicle in an efficient and swift manner.

[0020] The above embodiments have the advantageous effect of allowing the use of a linearized signal processing model, and thereby enabling the application of computationally efficient filtering algorithms to estimate the alignment angle on a road vehicle in an efficient and swift manner. Figure One Additional features of the application will be apparent from the disclosure of the application. BRIEF DESCRIPTION OF DRAWINGS

[0021] In the following, the embodiments herein will be described in more detail, only by way of example, with reference to the enclosed drawings, wherein:

[0022] Figure 1 A positioning and alignment of one radar sensor on the ego vehicle is schematically illustrated.

[0023] Figure 2 A road vehicle with a system comprising a controller for radar auto-alignment according to the methods described herein is schematically illustrated.

[0024] Figure 3 A feasible functional architecture for implementing radar auto-alignment according to the methods described herein is schematically illustrated.

[0025] Figure 4 A data processing arrangement for performing the various steps and processes described herein is schematically illustrated.

[0026] Figure 5 How the processing can be performed at a location remote from the road vehicle according to alternative embodiments is schematically illustrated. DETAILED DESCRIPTION

[0027] Some example embodiments of a method of determining alignment angles of two or more road vehicle 1 carried radar sensors 4 for a road vehicle radar auto-alignment controller 3 will be described below, a road vehicle system 2 comprising a controller 3 for radar auto-alignment according to said method, and a road vehicle 1 comprising such a system 2 and a computer program 12 for causing such a system 2 to perform said method.

[0028] Radar auto-alignment is the problem of unsupervised precise determination of the angle between the primary beam direction of a radar sensor 4 and the direction of the associated road vehicle 1. This angle varies due to installation tolerances in the factory, but there is also usually some drift due to wear and temperature changes when the radar sensor 4 is in operation, which must be continuously compensated.

[0029] The problem considered here is therefore to estimate the radar 4 alignment angle online in a computationally efficient way. The method presented here is also able to estimate longitudinal and lateral speed as well as side slip rate.

[0030] The presented method is not only used to compensate for the bias of the angle reported by the radar 4, but also for the calibration of vehicle dynamics ego-motion data, which are mainly longitudinal speed and lateral speed, for monitoring of the radar 4, for example from a safety point of view.

[0031] Thus, described herein is a method of determining alignment angles of two or more road vehicle 1 carried radar sensors 4 for a road vehicle radar auto-alignment controller 3 starting from an initially available coarse estimate of the alignment angle.

[0032] The initially available coarse estimate of the alignment angle can be obtained, for example, from known nominal installation angles of the vehicle carried radar sensors 4. These known nominal installation angles of the vehicle carried radar sensors 4 can be obtained, for example, from a road vehicle 1 related database comprising nominal parameters for the radar sensors 4, such as data about the installation alignment angle of the radar sensors 4.

[0033] The obtained signals from at least two radar sensors 4 are related to the detected range, azimuth and range rate. This is because the presented system of equations is underdetermined for one radar 4. According to the presented method, signals from at least two radars 4 are needed for a single instantaneous estimate of the alignment angle and the road vehicle 1 speed.

[0034] The detections are filtered 5 to determine detections from stationary targets. Filtering the detections to determine detections from stationary targets can be done, for example, by monitoring the longitudinal and lateral velocity and the side slip rate of the road vehicle 1 together with the obtained rate of change of distance of the detections. Thus, for example, the difference in the velocity of the detections projected in the azimuth direction can be measured by the radar 4 and combined with the ego vehicle 1 velocity to infer whether the detection is from a stationary target.

[0035] Alternatively, the detections can be filtered 5 to determine detections from stationary targets using the so-called Random Sample Consensus (RANSAC) technique or by running a target tracking and looking at the time effect.

[0036] RANSAC is an iterative method for estimating the parameters of a mathematical model from a given set of observed data that contains outliers. Thus, it can also be interpreted as an outlier detection method. RANSAC is a non-deterministic algorithm in the sense that it only produces reasonable results with a certain probability, which increases as more iterations are allowed. The basic assumption is that the data consists of "inliers", which means data whose distribution can be explained by some set of model parameters, although these parameters can be affected by noise, and "outliers", which means data that does not fit the model. Outliers can come, for example, from extreme values of noise or from erroneous measurements or incorrect assumptions about the interpretation of the data. RANSAC also assumes that given a set of inliers, which is usually small, there is a process that can estimate the parameters of a model that can best explain or fit these data.

[0037] A linearized signal processing model is derived from the determined detections from stationary targets, which model involves the alignment angle, the longitudinal and lateral velocity and the side slip rate of the road vehicle. The linearized signal processing model can be derived by performing a Taylor series expansion of a rotation matrix, which involves the alignment angle, the longitudinal and lateral velocity and the side slip rate of the road vehicle, as described below.

[0038] To derive the signal processing model of the work presented here, it is first considered that Figure 1 , Figure 1 The positioning and alignment of one radar sensor 4 on the ego vehicle 1 is schematically shown.

[0039] Here, v x and v y are the longitudinal and lateral velocity of the ego vehicle 1 at the rear axle center of the ego vehicle 1. The longitudinal distance from the rear axle center of the ego vehicle 1 to the position of the i-th radar is denoted by δ x，i , while the lateral distance is denoted by δ y，i . The alignment angle is denoted by βi and ego-vehicle velocity expressed in radar coordinate system are denoted as v x，i and v y，i respectively. Assuming planar motion, ego-vehicle velocity in radar coordinate system is obtained by the following.

[0040]

[0041] In the above equation, Ω z denotes the vehicle side slip rate. From the publication of Kellner et al. in 2014 IEEE International Conference on Robotics and Automation, "Instantaneous Ego-motion Estimation using Multiple Doppler Radars", pp 1592-1597, it is assumed that for the i-th radar, there are M i detections from stationary targets, we have the following model:

[0042]

[0043] Here, for the j-th detection, v j,i D is the Doppler velocity, also known as "range rate", and θ j,i is the azimuth angle in the sensor frame for the j-th detection.

[0044] To perform the estimation, first, we aim to compress the amount of data by computing the following velocity:

[0045]

[0046] The above computation relies on the Moore Penrose pseudo-inverse of a matrix to compute the estimate.

[0047] Note that this estimate depends on the ability to successfully filter the data for stationary detections, which will be kept as anchors. Typically, this filtering process is based on monitoring the vehicle dynamics ego-vehicle velocity together with the radar range rate. However, this filtering process can sometimes be difficult to make robust, and optionally one can therefore wish to consider the so-called RANSAC technique for deciding which detections originate from stationary objects.

[0048] To proceed, after some rearrangement, the model now reads as follows:

[0049]

[0050] To arrive at the proposed signal processing model, we will take the derivative w.r.t. β iThe linearized signal processing model, as described above, assumes that we have a rough estimate of the initial available alignment angle: β i = β i nom + ξ i In practice, this is a reasonable assumption because the nominal installation angle is known and we are interested in estimating a small residual alignment angle (e.g., on the order of 1 degree). The linearized model now looks like:

[0051]

[0052] For notational simplicity, the above only writes the equations for one radar.

[0053] A filtering algorithm can now be applied to this linearized model in order to estimate the alignment angle. The filtering algorithm used to estimate the alignment angle is suitably one of a Kalman filter, a least mean square filter, a recursive least squares filter, and a windowed least squares filter, or a filtering algorithm based on other signal processing algorithms suitable for utilizing the linearized signal processing model.

[0054] Kalman filtering (also known as linear quadratic estimation (LQE)) is an algorithm that uses a series of measurements observed over time, which contain statistical noise and other inaccuracies, and produces estimates of unknown variables that tend to be more accurate than estimates based on a single measurement alone by estimating a joint probability distribution over variables at each time frame.

[0055] The algorithm works in a two-step process. In the prediction step, the Kalman filter produces estimates of the current state variables along with their uncertainties. Once the result of the next measurement is observed, these estimates are updated using a weighted average, with estimates having higher certainty given greater weight. The algorithm is recursive. It can run in real-time using only the current input measurement and the previously computed state and its uncertainty matrix; no additional past information is needed.

[0056] The least mean square (LMS) algorithm is a class of adaptive filters that are used to mimic a desired filter by finding filter coefficients that are related to the least mean square of an error signal (i.e., the difference between the desired signal and the actual signal). This is a stochastic gradient descent method because the filter adapts based only on the error at the current time.

[0057] The recursive least squares (RLS) is an adaptive filtering algorithm that recursively finds the coefficients that minimize a weighted least squares cost function related to the input signal. The RLS exhibits extremely fast convergence. However, this benefit comes at the cost of high computational complexity, which, as noted previously, can be undesirable due to the potential limitations of the processing power onboard a road vehicle.

[0058] Windowed Least Squares (WLS) is a recursive variant of least squares, which aims at minimizing the squared error, but this is over a limited window, not just one iteration.

[0059] In a preferred embodiment, the linearized model is inserted into a linear Kalman filter machine. Note that the above system of equations is underdetermined for one radar 4. At least two radars 4 are needed for a single-instantaneous-estimate of the alignment angle and the road vehicle 1 speed.

[0060] The advantage of using Kalman filtering is that, compared to prior art solutions such as US 2017261599 (A1), which performs the estimation for one radar at a time, the alignment angle can be estimated simultaneously for several radar sensors 4. Thus, the Kalman filtering both performs an averaging over time in order to reduce the noise and obtain a stable angle, and simultaneously combines data from e.g. four radars 4a, 4b, 4c and 4d, making it faster and more efficient than other alternatives.

[0061] In order to reach the above linearized signal processing, as mentioned above, the amount of data should be compressed, otherwise one would have to deal with a non-linear model, which would be difficult, if not impossible, to handle in a road vehicle 1 due to potential limitations in processing power. The linearized signal processing model presented herein allows for simple real-time processing in a road vehicle 1, making it very useful.

[0062] Based on the estimated alignment angle is the generation of a signal suitable for causing the road vehicle 1 radar auto-alignment controller 3 to perform radar 4 offset compensation. For each radar sensor 4, such a signal can e.g. correspond to the difference between the assumed initial alignment angle and the actual estimated alignment angle.

[0063] As Figure 2 A road vehicle 1 system 2 comprising a controller 3 for radar 4 auto-alignment according to the method described herein, and a road vehicle 1 comprising a system 2 with such a controller 3 for radar 4 auto-alignment are also envisaged, as shown in

[0064] The system 2 can be an Advanced Driver Assistance System (ADAS), such as e.g. a system for assisting vehicle positioning, for providing collision warnings or avoidance or similar etc.

[0065] Figure 3A feasible functional architecture for implementing radar auto-alignment according to the methods described herein is schematically shown. Radar sensors (front left 4a, front right 4b, rear left 4c and rear right 4d) are arranged to provide range, azimuth and range rate to detected objects. A screening 5 is performed to find detections from stationary objects. A linearized processing model is subject to a Kalman filter 6 which provides alignment angle offsets for the four radars 4a, 4b, 4c and 4d. Radar offset compensation 7 is performed using the alignment angle offsets for the four radars 4a, 4b, 4c and 4d obtained from the Kalman filter 6 for producing an alignment compensated signal 10. Furthermore, since the Kalman filter 6 can also provide longitudinal velocity, lateral velocity and side slip rate of the ego vehicle 1, it is for example possible to use the longitudinal velocity provided by the Kalman filter 6 to estimate a scaling error 8 with respect to the longitudinal velocity provided from vehicle dynamics 9 to provide an improved longitudinal velocity signal 11.

[0066] Furthermore, it is herein envisaged a computer program 12, e.g. embodied on a non-transitory computer readable storage medium, the computer program 12 comprising program code for controlling a road vehicle 1 radar auto-alignment controller 3 to perform a process for road vehicle 1 radar auto-alignment, the process comprising the method described herein.

[0067] As Figure 4 indicated in the background art, the system 2 can comprise one or more data processing arrangements 16, each arrangement 16 comprising a processing module 17, said processing module 17 typically comprising at least one processing circuitry 14 comprising one or more processors and comprising at least one memory 18 storing instructions executable by the processing circuitry 14, e.g. a computer program 12 (software) comprising instructions for performing the various steps and processes described herein. It typically also comprises an I / O module 19 preparing for input of data to be processed and output of results of such processing.

[0068] Such computer program 12 can comprise instructions which, when executed by the processing circuitry 14, are configured to cause the road vehicle system 2 to perform the described method. The processing circuitry 14 can comprise a set of one or more processors (not shown). The carrier (not shown) can comprise a computer program 4 wherein the carrier can be one of an electronic signal, optical signal, radio signal, or computer readable storage medium.

[0069] The processing circuitry 14 can be arranged in the road vehicle 1 system 2, as Figure 2 indicated in the background art, or at a location remote from the road vehicle 1, e.g. a remote server 15, as Figure 5 indicated in the background art.

[0070] In embodiments where the processing circuitry 14 is arranged at a remote server 15, the method can be implemented, for example, by continuous streaming of data between the processing circuitry 14 and the road vehicle 1. Streaming is the delivery of content in real-time, i.e. a continuous stream of data, as it happens.

[0071] As Figure 4 As shown in Fig. 1, the system 2 or remote server 15 can comprise one or more data processing arrangements 16, each arrangement 16 comprising a processing module 17, which typically comprises at least one processing circuitry 14 comprising one or more processors and at least one memory 18 storing instructions executable by the processor, e.g. a computer program 12 (software), which comprises instructions for performing the various steps and processes described herein, and further comprising an I / O module 19, which prepares input data to be processed and output results of such processing.

[0072] The streaming of data between the road vehicle 1 and the processing circuitry 14 located at such a remote server 15 (cloud) and back to the road vehicle system 2 comprising the controller 3 for radar auto-alignment of the road vehicle 1 can further comprise a communication network connected to the remote server 15, e.g. as illustrated by arrow 20. Such a communication network 20 represents one or more mechanisms by which the road vehicle 1 can communicate with the remote server 15. Thus, the communication network 20 can be one or more of a variety of wireless communication mechanisms, including wireless (e.g., radio frequency, cellular, satellite, and microwave communication mechanisms) and any desired combination of network topologies. Exemplary communication networks include wireless communication networks, e.g. using Bluetooth, IEEE 802.11, LTE, 5G, etc.

[0073] The combination of all the above factors contributes to an improved method of determining alignment angles of two or more road vehicles carrying radar sensors in a computationally efficient manner, which is suitable for on-board use in a road vehicle radar auto-alignment controller.

[0074] Many modifications and other embodiments of the present invention set forth herein will come to mind to one skilled in the art to which the invention pertains having the benefit of the teachings presented in the foregoing description and the associated drawings. Therefore, it is to be understood that the invention is not to be limited to the specific embodiments disclosed and that modifications and other embodiments are intended to be included within the scope of the appended claims. Without intent to limit the scope of the application, exemplary methods and their related components are generally described as follows.

[0075] Moreover, while the foregoing description and the associated drawings set forth exemplary embodiments, it will be understood and appreciated that the subject matter is not limited to those examples. Instead, many alternatives, modifications, permutations, and variations of the examples disclosed herein are possible. For example, while the above description and associated drawings set forth exemplary embodiments, it will be appreciated that the features and / or functions of one or more of the above-described embodiments can be combined with features and / or functions of one or more of the other embodiments. As such, the above description and associated drawings should not be taken as limiting the claimed application to any single embodiment or group of embodiments. Rather, the scope of the application should be determined with reference to the appended claims and their equivalents.

[0076] Where an advantage, benefit, or solution to a problem is described herein, it will be appreciated that such advantages, benefits, and / or solutions can be applicable to some example embodiments but not necessarily all example embodiments. As such, any advantages, benefits, or solutions described herein should not be taken as crucial, required, or necessary to all embodiments or embodiments claimed herein. Although specific terms are employed herein, they are used in a generic and descriptive sense only and not for purposes of limitation.

Claims

1. A method of determining an alignment angle of radar sensors (4) carried by two or more road vehicles (1) starting from an initially available coarse estimate of the alignment angle from an alignment angle perspective, the alignment angle being defined as an angular difference with respect to a lateral velocity direction of the road vehicle, the method comprising: obtaining signals related to detected range, azimuth and range rate from at least two radar sensors (4); screening (5) the detections to determine detections from stationary targets; deriving a linearized signal processing model from the determined detections from stationary targets, the model involving the alignment angle, longitudinal and lateral velocity and side slip rate of the road vehicle, wherein the signal processing model is linearized with respect to the alignment angle in case of a coarse estimate of the alignment angle initially available; applying (6) a Kalman filter algorithm to the linearized signal processing model to estimate the alignment angle; generating the signal suitable for causing a road vehicle radar auto-alignment controller to perform radar offset compensation based on the estimated alignment angle.

2. The method according to claim 1, wherein the method further comprises obtaining the initially available coarse estimate of the alignment angle from a known nominal mounting angle of the radar sensors (4) carried by the vehicle.

3. The method according to any one of claims 1 to 2, wherein the method further comprises screening the detections to determine detections from stationary targets by monitoring the longitudinal and lateral velocity and side slip rate of the road vehicle (1) together with the obtained range rate of the detections.

4. The method according to any one of claims 1 to 2, wherein the method further comprises screening the detections to determine detections from stationary targets using RANSAC techniques or by running target tracking and searching for temporal effects.

5. The method according to any one of claims 1 to 2, wherein the method further comprises deriving the linearized signal processing model by performing a Taylor series expansion of a rotation matrix involving the alignment angle.

6. A road vehicle (1) system (2) comprising a road vehicle radar auto-alignment controller (3) for performing radar offset compensation based on a signal generated according to the method of any one of claims 1 to 5.

7. A road vehicle (1) comprising a road vehicle system (2) according to claim 6.

8. A computer readable storage medium having stored thereon a computer program comprising program code configured to perform the method of any one of claims 1 to 5 when the program code is executed by processing circuitry (14).

9. A computer program product comprising a computer program (12) comprising instructions configured to cause the road vehicle (1) system (2) to perform the method of any one of claims 1 to 5 when the instructions are executed by processing circuitry (14).

Images