A method for modeling sensor uncertainty in central-level tracking architectures.

Abstract

A method for modeling sensor uncertainty in central-level tracking architectures. This patent relates to a method for modeling sensor uncertainty in a central level tracking architecture system that fuses multiple sensor detections for at least one sensor object. The detections obtained by the sensors correspond to at least one dimension of the at least one sensor object. A predetermined association gate minimum threshold is set for each dimension for each sensor by setting a predetermined amount of sensor detection error for each dimension, and an increase of each sensor detection error is modeled as a function of distance based on the predetermined amount of sensor detection error. An association gate is calculated by increasing each sensor detection error that exceeds the predetermined association gate minimum threshold according to the modeling. A covariance matrix associated with each sensor detection for each sensor is calculated. Finally, the multiple covariance matrices corresponding to all sensors are stored and made available to the central level tracking system for fusing the multiple sensor detections.

Description

[Technical field]

[0001] The present invention relates to fusing detections from different types of sensors in a multi-sensor tracking system, and in particular to modeling sensor uncertainty in a central level tracking system architecture.

[0002] Terms used in the present invention A measurement, also called a detection or sensor detection, is data corresponding to a value received in a current observation sequence by a sensor in a multiple sensor tracking system.

[0003] A sensor object, also called a target object, is a real-world object that is detected by one or more sensors.

[0004] A fusion object is a sensor object for which measurement fusion was performed in the immediately preceding observation sequence.

[0005] The target position for each sensor object is the estimated position of the respective sensor object in the current observation sequence. [Background technology]

[0006] It is known from the prior art that fusing detections from different types of sensors in a multi-sensor tracking system has the advantage that sensor object coverage is wider and the estimation of the subsequent position of the target object is more accurate.

[0007] The target object may be a single target object or multiple targets.

[0008] In a central level tracking architecture, all detections acquired by the sensors are transmitted to a central tracking system which keeps track of each detection's corresponding track.

[0009] The theoretical advantage of a central level tracking architecture is that it makes the most of all the information contained in the detection.

[0010] One known problem in multi-sensor estimation is the so-called data association problem, which basically concerns knowing which existing fusion objects should be associated with which new measurements coming from a particular one of the sensors.

[0011] As pointed out in Non-Patent Document 1, "The large number of measurements made need to be correctly associated with the underlying states being observed. This is the data association problem. Data association problems include the problems of validating the data (e.g., ensuring that it is not erroneous or does not result from clutter), associating the correct measurements with the correct states (especially in multi-target tracking tasks), and initializing new tracks or states as necessary. While conventional tracking is actually concerned with uncertainties in the measurement locations, data association is concerned with uncertainties in the measurement origins."

[0012] Disadvantages of the Prior Art One known method of dealing with sensor uncertainty is based on "data-driven modeling" of the sensor error, i.e., performing a measurement run, extracting data from the sensor, and using fitting techniques to determine the uncertainty.

[0013] This approach has two disadvantages: on the one hand, it is generally expensive since special drives and measurement sessions are required to collect data on the field of view of the sensor, and on the other hand, it results in uncertainty maps that require large memory to store, which is problematic for embedded systems since it reduces their performance.

[0014] Another approach known from the prior art literature is to directly use the information provided by the sensor. This approach is only applicable if the sensor provides sufficient statistical information regarding the data it provides. This is also a disadvantage since in practice this is usually not the case when the sensor manufacturer does not provide this type of information. [Prior art documents] [Non-patent literature]

[0015] [Non-Patent Document 1] Hugh Durrant-Whyte, “Multi Sensor Data Fusion”, Australian Center for Field Robotics, The University of Sydney, 2006. Summary of the Invention [Problem to be solved by the invention]

[0016] The problem that the present invention aims to solve is to model the sensor uncertainties more accurately in the central level tracking architecture, i.e. to define association gates for different sensors faster and cheaper, which leads to improved tracking performance and better definition of association gates for different sensors.The objective of the present invention is to provide decision makers with better tools for describing the measurement errors due to sensors, which is useful for the fusion of sensor detections. [Means for solving the problem]

[0017] To achieve this object, the inventors have provided in a first aspect of the present invention a computer implemented method for modelling sensor uncertainty in a central level tracking architecture system fusing multiple sensor detections for at least one sensor object, each of the at least one sensor object being located at a respective distance from each of the sensors, the distance being calculated with respect to a principal axis of each of the sensors, the method comprising: The following steps are performed during the current observation sequence: S1.1. Obtaining and storing a sensor detection of at least one sensor object for each sensor, each sensor detection corresponding to at least one dimension of the at least one sensor object, each sensor detection including a respective sensor detection error; S1.2. Setting a predetermined association gate minimum threshold for each dimension for each sensor by setting a predetermined amount of sensor detection error for each dimension for each sensor; S1.3. Modeling the increase in each sensor detection error for each sensor using a predetermined amount of sensor detection error as a function of distance; S1.4. Calculating an association gate for each sensor detection corresponding to each dimension for each sensor by increasing each sensor detection error that exceeds a predetermined association gate minimum threshold according to the modeling; S1.5. Calculating, for each sensor, a covariance matrix associated with each sensor detection using a scaling technique; and S1.6. We have devised a method comprising the steps of storing multiple covariance matrices corresponding to all sensors and making the multiple covariance matrices available to a central level tracking system for fusing multiple sensor detections of a current observation sequence with at least one fusion object of a previous observation sequence, each at least one fusion object corresponding to each at least one sensor object.

[0018] Each covariance matrix indicates the shape of the uncertainty for its corresponding sensor corresponding to each dimension.

[0019] In a second aspect of the present invention, there is provided data processing hardware having at least one computer processing unit core, comprising at least one volatile memory RAM and at least one non-volatile memory ROM, characterized in that the data processing hardware is configured to execute a computer implemented method for modelling sensor uncertainty according to any preferred embodiment.

[0020] In a third aspect of the invention there is provided a computer program comprising instructions which, when executed by data processing hardware, cause the data processing hardware to perform the steps of the computer implemented method according to any preferred embodiment. Effect of the Invention

[0021] Advantages of the invention The main advantages of the present invention are as follows: The method of the present invention allows faster and cheaper estimation of the measurement uncertainty provided by the sensors compared to the prior art. Improved uncertainty estimation provides decision makers with more tools to improve the correction of sensor-induced errors and better exploits the characteristics of each sensor in the fusion.

[0022] The method of the present invention is robust, does not require large memory or processing power, does not require extensive statistical information from the sensor manufacturer, and can easily fit into embedded systems.

[0023] The method of the invention can be used in any technical field in which a sensor object is moving and at least one of different types of sensors detects the movement of the sensor object, for example in the automotive industry or the earthmoving or construction industry. [Brief description of the drawings]

[0024] [Figure 1] FIG. 1 illustrates the principle of calculating the association gate using polar coordinate representation, and in order to facilitate understanding of the invention, the position of at least one fusion object FOk is represented two-dimensionally in a plane. [Diagram 2] FIG. 2 illustrates the principle of calculating the association gate using a Cartesian coordinate representation, where the position of at least one fusion object FOk is represented two-dimensionally in a plane to facilitate understanding of the invention. [Figure 3A]FIG. 3A illustrates the principle of creating an association gate for one radar and one camera of a vehicle using a Cartesian coordinate representation. [Figure 3B] FIG. 3B illustrates the principle of creating an association gate for one radar and one camera of a vehicle using a Cartesian coordinate representation. [Figure 4] FIG. 4 shows the matching ellipsoids of the covariance matrices for the camera and the radar, and for ease of understanding the invention, the position of at least one fusion object FOk is displayed two-dimensionally in a plane. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0025] The method according to the invention is used in a central level tracking architecture system for fusing sensor detections acquired by sensors for a single sensor object or for multiple sensor objects.

[0026] The method of the present invention can be applied to any land vehicle equipped with a sensor, which is referred to as the ego vehicle, for example a car, a truck, a trailer roll, a bus, a crane, an earth moving machine, a tractor, a motorcycle, a two-wheeler, etc.

[0027] The ego vehicle may be autonomous or manned.

[0028] For the sake of brevity, throughout the detailed description and examples of the embodiment, the distance d k A single sensor object O located at k and the distance d k is calculated for each principal axis of each sensor, and each sensor detection is the sensor object O k A single sensor object O k In each case where is mentioned, a single sensor object O k Applying mutatis mutandis the teachings disclosed for all sensor objects O k It should be understood that this applies to

[0029] The method of the present invention requires the use of at least two sensors since there is no fusion in sensor detection obtained with a single sensor.

[0030] For simplicity, throughout this specification, the term "sensor" refers to a sensor of the ego vehicle selected to provide sensor detection for implementing the method of the present invention in any embodiment. The ego vehicle may also be equipped with other sensors not related to the present invention.

[0031] The sensor object O of the present invention k The host vehicle is generally in motion, but may also be temporarily stopped, for example while waiting at a traffic light.

[0032] Each sensor object O k has its own unique shape depending on the technology, the object position and the sensor mounting location.

[0033] The sensor is connected to each sensor object O k The movement of each sensor object O located within the field of view FOV of each sensor is tracked. k The method of the present invention measures at least one dimension of a sensor object O located simultaneously in the field of view FOV of two or more sensors. k Applies to.

[0034] Measurements are made by sensors according to the specifications of each sensor's respective manufacturer.

[0035] Regardless of the application of the present invention, the sensor measurements are transmitted to one or various fusion units in a central level tracking system architecture for sensor fusion.

[0036] The sensor measurements are acquired by data processing hardware which performs all steps of the present invention. Acquired sensor measurements, also referred to as "detections" or "sensor detections", refer to the data of values ​​acquired by each sensor, which are then processed by the data processing hardware in the method steps as disclosed.

[0037] The method according to the invention comprises six steps, all of which are executed by the data processing hardware on the current observation sequence.

[0038] In the immediately preceding observation sequence, a particular sensor object O k At least one fusion object FO corresponding to k should be understood as having been calculated.

[0039] In the first step, each sensor object O k , and each sensor detection is stored for at least one sensor object O k Each sensor detection includes a respective sensor detection error.

[0040] In the second step, the distance d k , a predetermined association gate minimum threshold is set for each dimension for each sensor by setting a predetermined amount of sensor detection error for each dimension for each sensor as a function of .

[0041] In the third step, the distance d k The increase in each sensor detection error is modeled for each sensor with a predetermined amount of sensor detection error as a function of .

[0042] Modeling of uncertainty is done for each sensor and each sensor object. k Independently, that is, the modeling parameters are For sensors, Same parameters for all sensors or may be different parameters for some or all of the sensors, Sensor object O k Regarding All sensor objects O k For the same parameters or Some or all of the sensor objects O k may be any of the different parameters.

[0043] Uncertainty modeling is performed using information provided by the sensor manufacturer, from which the standard deviation to be used in describing the sensor error is estimated.

[0044] Then, in a fourth step, an association gate is calculated for each sensor detection corresponding to each dimension for each sensor by multiplying each sensor detection error that exceeds a predetermined association gate minimum threshold according to the modeling.

[0045] The upper limit of the increase is defined by the measurement limits of each sensor as stated in its corresponding technical specification.

[0046] The association gate, also called the verification gate, is a fusion gate that detects each sensor in the current observation sequence and the fusion object F k Each sensor detection is the observation region around the corresponding sensor object O k Indicates the target position.

[0047] The size of the gate is set using a criterion available from the prior art literature based on an assessment of the probability of false association versus the probability of missing association, i.e., using the chi-square distribution. The size of the gate is heavily influenced by the size and shape of the sensor errors modeled in steps 2 and 3, although indirectly, through the fusion algorithm.

[0048] As is well known from the prior art literature, models of sensor uncertainty are important to be able to best use the measurements.

[0049] An extreme example would be the fusion of detections from two sensors, one with very good position measurements in the X direction and very bad position measurements in the Y direction, and the second with the opposite characteristics, i.e. very good in the Y direction and very bad in the X direction. The fusion method used is a Kalman filter (see for example https: / / en.wikipedia.org / wiki / Kalman_filter), and the resulting position estimate often uses the X position of the first sensor and the Y position of the second sensor, assuming that they are based on the relevant equations, which are assumed to be known here. To obtain this result from the fusion algorithm, and not limited to the Kalman filter case, it is important to feed the fusion algorithm with an accurate description of the measurement errors. The same description also influences the association gate definition, due to the fact that the more accurate the sensor values ​​are, the more strict the association criteria and the less accurate the weaker the association criteria. This is reflected when the chi-squared distribution is used in the fifth step above.

[0050] In a fifth step, for each sensor a scaling technique is used to calculate the covariance matrix associated with each sensor detection.

[0051] The covariance matrix is ​​the distance d in the field of view of the sensor k is a function for each dimension that can be calculated for each association gate by scaling, which depends on: Scaling can be done using a variety of formulas, a non-limiting example is linear interpolation for distance and angle.

[0052] The covariance matrix indicates the shape of the uncertainty for each dimension for its corresponding sensor.

[0053] In a sixth step, multiple covariance matrices corresponding to all sensors are stored, and the multiple sensor detections of the current observation sequence are fused with at least one fused object F(O) of the immediately preceding observation sequence. kand making available to the central level tracking system a plurality of covariance matrices for fusing with each at least one fused object F.O. k Each sensor object O k Corresponds to.

[0054] The overlap of the association gate or each of the association gates of the fusion object in the immediately preceding observation sequence with the association gate or each of the association gates of each sensor detection in the current observation sequence is verified, thereby determining whether each sensor detection is one or more fusion objects FO k This is taken as an indication that the sensor detection fusion may be fused with the . The rules of validation and sensor detection fusion are outside the scope of the present invention since the present invention merely provides better tools for determining the uncertainty of the sensor detection.

[0055] The disclosed method of the invention has the advantage of being a simple, yet quantitative, faster and cheaper method for estimating the measurement uncertainty provided by a sensor compared to the prior art, since the covariance matrix output by the method as the uncertainty formula is based on theory and on data available in the sensor documentation of the sensor manufacturer, so the calculation time is shorter than that of the prior art, and on the other hand does not require long, complex and expensive measurement actions with the sensor accuracy, so the covariance matrix and the method are cheaper than the prior art methods.

[0056] Improving the uncertainty description provides decision makers with more tools to improve the use of sensor detection to exploit each sensor's features faster and cheaper in fusion.

[0057] The method of the present invention is robust and can easily fit into embedded systems without requiring large memory or processing power, since the calculation of the covariance matrix does not require large memory or processing power, and by using the method of the present invention, sufficient statistical information from the sensor manufacturer is no longer required, since a simplified set of information provided in the sensor data sheet is usually sufficient to calculate a sufficient covariance matrix to provide a purposive description of the sensor fusion, making the present invention particularly advantageous for sensors where the sensor manufacturer does not provide sufficient information about the errors.

[0058] In a preferred embodiment, the method further comprises a sub-step of the fifth step. In a fifth step, each covariance matrix is ​​graphically represented using a matching ellipsoid proportional to the size of the respective covariance matrix and proportional to the size of the association gate corresponding to each dimension.

[0059] Each matching ellipsoid indicates the shape of the uncertainty of its corresponding sensor for each dimension, and the larger the size of the ellipsoid, the correspondingly larger the degree of detection uncertainty of each sensor.

[0060] By graphically displaying multiple covariance matrices as matching ellipsoids, the present invention has the further advantage of enabling a decision maker to have a graphical overview as to which sensors and / or which regions of the sensor's field of view FOV require action by the decision maker to reduce uncertainties. An example implementation and FIG. 4 show such an overview diagrammatically.

[0061] The method of the present invention, with or without the graphical representation of multiple covariance matrices, is applicable to a variety of numbers of dimensions, of which three preferred embodiments are presented below.

[0062] In one preferred embodiment, only one dimension, i.e. the sensor object O k For simplicity of graphic representation, all figures are based on the sensor object O k 4 shows the application of the method to the position of

[0063] In another preferred embodiment, not shown, the two dimensions are k For simplicity, the position of the sensor object O k The velocity of the sensor object O relative to the ego-vehicle, called k There exists a relative velocity of

[0064] In another preferred embodiment, not shown, the three dimensions are k There exists a position, velocity and acceleration.

[0065] The choice of the type and number of dimensions takes into account the specific needs for estimating the corresponding variables. For example, if the ego vehicle is an earthmoving machine and it is known that this type of machine moves very slowly compared to passenger cars, then the sensor object O k It is not necessary to use acceleration, and in some circumstances, velocity.

[0066] In contrast, if the ego-vehicle is a road vehicle, and in particular if the ego-vehicle is moving above a certain speed, e.g. above 60 km / h, and the sensor uncertainty needs to be reduced, the sensor object O k It is convenient to select the position, velocity and acceleration of for use in the method.

[0067] In a preferred embodiment, the sensors are of the same type, i.e. radar sensors, cameras, ultrasonic sensors, lidar sensors, etc.

[0068] In another preferred embodiment, the sensors are of two or more types, i.e., radar sensor and camera, radar sensor and ultrasonic sensor, radar sensor and lidar sensor, camera, radar sensor and ultrasonic sensor, etc.

[0069] The number and combination of sensor types is defined by the existing configuration of sensors on the ego vehicle and the sensor selection made by the decision maker.

[0070] In one example, not shown, several ego-vehicles are equipped with a single type of sensor, for example a camera, in which case some or all of the cameras are used for application of the method of the invention.

[0071] In another example, not shown, the other host vehicle is equipped with three types of sensors: a camera, a radar sensor and an ultrasonic sensor.

[0072] In all instances, the method may be applied to all sensors in a vehicle or only a portion of the sensors.

[0073] Using the method of the present invention on only some of the sensors of the vehicle can be useful, for example, if a decision maker knows that there is a need to reduce the uncertainty of sensors detecting objects from certain areas of the vehicle, such as blind spots.

[0074] In another preferred embodiment, the sensor detection is obtained by at least one camera and at least one radar. This embodiment is shown in the implementation example and illustrated in Figures 3A, 3B and 4.

[0075] By allowing multiple combinations of numbers and types of sensors and multiple combinations of numbers and types of dimensions measured by each sensor, the present invention has the advantage of being able to adapt to each particular need for reducing the sensor uncertainty of the vehicle and to adapt to the existing configuration of sensors of the vehicle.

[0076] The present invention is independent of the reference system, ie, polar or Cartesian.

[0077] All calculations made in any step of the method according to any preferred embodiment can be done in either polar or Cartesian coordinates.

[0078] If necessary, the results of any step of the method, and the results of the method itself, can be converted from polar coordinates to Cartesian coordinates and vice versa.

[0079] The choice of whether to use polar or Cartesian coordinate representation is a matter of convenience, depending on which reference system simplifies the calculations and which reference system is used for fusion.

[0080] For example, assume that the tracking system is configured to use a Cartesian reference frame, while the radar measurement uncertainty is more simply described in polar coordinates since the global uncertainty is naturally partitioned into radial and angular uncertainties. For this example, the primary reference system is a Cartesian reference frame, and the radar measurement uncertainty is transformed from polar coordinates to Cartesian coordinates.

[0081] FIG. 1 shows diagrammatically the principle of calculating the association gate in polar coordinates for an embodiment in which there is only one dimension, namely the position. From FIG. 1, it can be seen that the radial distance is the distance d k It can be seen that the azimuth angle is the angle θ between the principal axis of the sensor and the reference direction of each sensor. Thus, the position p illustrated in FIG. k ,θ).

[0082] Figure 2 shows, for the same embodiment, a schematic of the same calculation of the association gate in Cartesian coordinates x,y, where for simplicity the reference direction is the y axis. From Figure 2 it can be seen that the position p is written as p(x,y).

[0083] In a second aspect of the present invention, there is provided data processing hardware having at least one computer processing unit core, comprising at least one volatile memory RAM and at least one non-volatile memory ROM, characterized in that the data processing hardware is configured to execute a computer implemented method for modelling sensor uncertainty according to any preferred embodiment.

[0084] In a preferred embodiment, the data processing hardware is a processor separate from the sensor processing device and communicates with the sensor processing device via a communications protocol.

[0085] In another preferred embodiment, data processing hardware is included in one or more sensor processing units and performs the normal tasks of the sensor processing unit in addition to the methods of the present invention.

[0086] In a third aspect of the invention there is provided a computer program comprising instructions which, when executed by data processing hardware, cause the data processing hardware to perform the steps of the computer implemented method according to any preferred embodiment.

[0087] Example implementation 3A, 3B and 4, a preferred embodiment is shown generally using two types of sensors in an example implementation: radar and camera.

[0088] For ease of understanding, a single sensor object O k A single fusion object FO corresponding to k is shown, and a sensor object O k Only the position p(x,y) of is shown.

[0089] 3A and 3B show the principle of creating an association gate for one radar and one camera of the ego vehicle.

[0090] In FIG. 3A, the situation after acquiring radar and camera sensor detections is shown diagrammatically. Each ellipse represents the area of ​​the fused object FO given the received measurements. k With about 68% confidence, the sensor object O k Describe the areas where this may be the case.

[0091] The fusion object FO of the previous observation sequence k The problem is to fuse two sensor detections of the current observation sequence into an object FO. k The question is whether or not to combine it with

[0092] The result of the association gate augmentation for both sensors is shown diagrammatically in Figure 3B, where the association gates are shown as two rectangles, and Figure 3B also shows the matching ellipsoid representation for two covariance matrices, one for the radar and the other for the camera.

[0093] Generally, the increase in a radially located associated gate, in this case a radar associated gate, is more significant in length than in width.Similarly, the increase in a laterally located associated gate, in this case a camera associated gate, is generally more significant in width than in length.

[0094] fused object FO k The overlap of the association gates with the fusion object FO is used to verify the sensor detection of the current observation sequence. k is taken as an indication that the

[0095] The overlap is evaluated using formulas that calculate distance measures between distributions, such as the Mahalanobis distance or the Kullback-Leibler divergence.

[0096] Based on the principles shown in Figures 3A and 3B and Figure 4, an example is shown in which the host vehicle is a passenger car or bus having four side radars provided at a front right angle, a front left angle, a rear right angle and a rear left angle, where the left and right are considered with respect to the direction of movement toward the positive side of the x-axis, and three cameras, one facing forward and two facing to the sides, i.e., the three cameras facing to the side left and the side right.

[0097] Moreover, the limit lines of the field of view FOV of the radar sensor are indicated by FR, FL, RR, and RL in FIG. 4, while the limit line of the FOV of the camera is indicated by C in FIG.

[0098] Figure 4 shows the matching ellipsoids of the two types of covariance matrices. covariance matrices CR for the camera and the forward-left and forward-right radars, where the two ellipsoids are perpendicular to each other; For all other parts of the field of view FOV, the ego-vehicle radar covariance Rc for the two radars and the camera covariance Cc for the two cameras, with the two ellipsoids one inside the other.

[0099] From Figure 4 it can be seen how the sensor uncertainty generally increases as distance and angle increase. Exactly how this occurs is related to the characteristics of the particular sensor as described by the sensor documentation and is reflected in the coefficients of the covariance matrix.

[0100] While the detailed description of the method and system has been disclosed in terms of preferred embodiments, it will be appreciated by those skilled in the art that various modifications and alterations of the present invention will be apparent to those skilled in the art without departing from the essential scope of the teachings of the present invention and without diminishing its advantages. Accordingly, such modifications and alterations are intended to be within the scope of the appended claims. This application relates to the invention described in the claims, but also includes the following as other aspects. 1. At least one sensor object (O k 1. A computer-implemented method for modeling sensor uncertainty in a central level tracking architecture system that fuses multiple sensor detections for each of the at least one sensor object (O k ) is the distance (d k ) and the distance (d k ) is calculated with respect to the principal axis of each of the sensors, The following steps are performed during the current observation sequence: S1.1. For each said sensor, the at least one sensor object (O k ) , each sensor detection being a signal of the at least one sensor object (O k ) corresponding to at least one dimension of the vector, each sensor detection including a respective sensor detection error; S1.2. Setting a predetermined association gate minimum threshold for each dimension for each of said sensors by setting a predetermined amount of sensor detection error for each dimension for each of said sensors; S1.3. The distance (d k modeling an increase in each sensor detection error for each said sensor using said sensor detection error predetermined amount as a function of S1.4. Calculating an association gate for each sensor detection corresponding to each dimension for each said sensor by increasing each sensor detection error that exceeds said predetermined association gate minimum threshold according to modeling; S1.5. Calculating, for each said sensor, a covariance matrix associated with each sensor detection using a scaling technique; and S1.6. Store multiple covariance matrices corresponding to all of the sensors and merge the multiple sensor detections of the current observation sequence with at least one fusion object (FO) of the immediately preceding observation sequence. k making the plurality of covariance matrices available to the central level tracking system for fusing with each of the at least one fused object (FO k ) is each of said at least one sensor object (O k ) Each of the covariance matrices indicates the shape of the uncertainty of the corresponding sensor for each dimension. A method comprising: 2. S2.5. Graphically displaying a matching ellipsoid for each said covariance matrix in proportion to the size of each said covariance matrix and in proportion to the size of the association gate corresponding to each dimension. 2. The method according to claim 1, further comprising: 3. The at least one sensor object (O k 3. The method according to claim 1 or 2, carried out for one dimension being the position of 4. The at least one sensor object (O k 4. The method according to claim 3, performed for a further dimension, which is the velocity of 5. The at least one sensor object (O k 5. The method according to claim 4, performed for a further dimension being the acceleration of 6. 6. The method according to any one of 1 to 5 above, wherein said at least one dimensional measurement is performed by acquiring sensor detections by sensors of the same type. 7. 6. The method according to any one of 1 to 5 above, wherein said at least one dimensional measurement is performed by acquiring sensor detections by sensors of different types. 8. 8. The method according to claim 7, wherein the sensor detection is obtained by at least one camera and at least one radar. 9. 9. A method according to any one of 1 to 8 above, wherein one or more steps of the invention are displayed in either polar or Cartesian coordinates, and the results of each step can be converted from polar coordinates to Cartesian coordinates or from Cartesian coordinates to polar coordinates, respectively. 10. 10. A data processing hardware having at least one computer processing unit core, at least one volatile memory RAM and at least one non-volatile memory ROM, characterized in that the data processing hardware is configured to execute the computer-implemented method for modeling sensor uncertainty according to any one of 1 to 9 above. 11. The data processing hardware according to claim 10, which is a processing device separate from the sensor processing device and communicates with the sensor processing device via a communication protocol. 12. 11. Data processing hardware according to claim 10 included in one or more sensor processing devices. 13. A computer program comprising instructions which, when executed by data processing hardware as defined in any one of claims 10 to 12 above, cause the data processing hardware to execute the steps of the computer-implemented method as defined in any one of claims 1 to 9 above. [Explanation of symbols]

[0101] Vehicle FO k fused object d k distance Radial distance d k and azimuth angle θ fusion object FO k Polar coordinates of p position x,y fused object FO k Cartesian coordinates of FL Front left radar field of view FOV limit line FR Front right radar field of view FOV limit line RL Left rear radar field of view FOV limit line RR Right rear radar field of view FOV limit line C. The limit of the forward-facing camera's FOV Rc Radar-radar covariance Cc Camera-camera covariance CR Camera-radar covariance

Claims

1. At least one sensor object (O ｋ 1. A computer-implemented method for modeling sensor uncertainty in a central level tracking architecture system fusing multiple sensor detections for each of said at least one sensor object (O ｋ ) is the distance (d ｋ ) and the distance (d ｋ ) is calculated with respect to the principal axis of each of said sensors, The following steps are performed in the current observation sequence: S1.

1. For each said sensor, the at least one sensor object (O ｋ ) , each sensor detection being associated with at least one sensor object (O ｋ ) corresponding to at least one dimension of the vector matrix, each sensor detection including a respective sensor detection error; S1.

2. Setting, for each said sensor, a predetermined association gate minimum threshold for each dimension by setting, for each said sensor, a predetermined amount of sensor detection error for each dimension as a function of said distance (d k ); S1.

3. The distance (d ｋ modeling an increase in each sensor detection error for each said sensor using said sensor detection error predetermined amount as a function of S1.

4. Calculating an association gate for each sensor detection corresponding to each dimension for each said sensor by multiplying each sensor detection error that exceeds said predetermined association gate minimum threshold according to modeling; S1.

5. Calculating for each of the sensors a covariance matrix associated with each sensor detection using a scaling technique, in particular a linear interpolation of the distance (d k ) and the angle (θ) between the sensor's main axis and the reference direction of each sensor, the covariance matrix being a function for each dimension depending on the distance (d k ) in the field of view of the sensor, calculable by scaling of each of the association gates; and S1.

6. Calculate covariance matrices as described in step S1.5, store the covariance matrices corresponding to all the sensors, and merge the sensor detections of the current observation sequence with at least one fusion object (FO) of the immediately preceding observation sequence. ｋ making the plurality of covariance matrices available to the central level tracking architecture system for fusing with each of the at least one fusion object (FO), ｋ ) is each of said at least one sensor object (O ｋ ) corresponding to the step of Each of the covariance matrices indicates the shape of the uncertainty of the corresponding sensor for each dimension. A method comprising:

2. S2.

5. Graphically displaying a matching ellipsoid for each said covariance matrix in proportion to the size of each said covariance matrix and in proportion to the size of the association gate corresponding to each dimension. The method of claim 1 further comprising:

3. The at least one sensor object (O ｋ 3. The method of claim 1 or 2, performed for one dimension being the position of

4. The at least one sensor object (O ｋ 4. The method of claim 3, wherein the method is performed for a further dimension, the velocity of

5. The at least one sensor object (O ｋ 5. The method of claim 4, wherein the method is performed with respect to a further dimension, which is acceleration of the

6. The method according to any one of claims 1 to 5, wherein said measurement of at least one dimension is performed by acquiring sensor detections with sensors of the same type.

7. The method according to any one of claims 1 to 5, wherein said measurement of at least one dimension is performed by acquiring sensor detections with sensors of different types.

8. The method of claim 7, wherein the sensor detection is obtained by at least one camera and at least one radar.

9. 9. A method according to any one of claims 1 to 8, wherein one or more steps of the invention are displayed in either polar or Cartesian coordinates, and the results of each respective step can be converted from polar coordinates to Cartesian coordinates or from Cartesian coordinates to polar coordinates, respectively.

10. A data processing hardware having at least one computer processing unit core and comprising at least one volatile memory RAM and at least one non-volatile memory ROM, characterized in that the data processing hardware is configured to execute a computer-implemented method for modelling sensor uncertainty according to any one of claims 1 to 9.

11. The data processing hardware according to claim 10 , wherein the processing device is a processing device separate from the sensor processing device and communicates with the sensor processing device using a communication protocol.

12. 11. Data processing hardware according to claim 10, included in one or more sensor processing devices.

13. A computer program comprising instructions which, when executed by data processing hardware according to any one of claims 10 to 12, cause the data processing hardware to carry out the steps of the computer-implemented method according to any one of claims 1 to 9.

Images