Method for tracking at least one object with at least one detection device, detection device and vehicle having at least one detection device

Abstract

The invention relates to a method for tracking at least one object (18) by means of at least one detection device, in particular of a lidar system, in particular of a vehicle, in which a plurality of target data are determined by using electromagnetic scanning signals, each target data comprising at least one position value, the position value characterizes a position of an object target of the at least one object (18) in the at least one predefined reference system. At least one piece of boundary data (x ', y', z ', alpha, l, w, h), which characterizes the shape of the at least one object (18), is determined from at least a portion of the target data. At least one object (18) is tracked by determining the development of at least one boundary data (x ', y', z ', alpha, l, w, h). Wherein in order to determine at least one boundary data (x ', y', z ', alpha, l, w, h), at least a portion of the target data is grouped into clusters, at least one weighting value (V) of the target data of the at least one cluster is determined from the target data of the respective cluster, the at least one boundary data (x', y ', z', alpha, l, w, h) is determined from the weighted target data (phi *, D *, I *, V *), v *) is determined from the respective target data weighted by at least one weighting value (V) of the cluster to which the target data belongs.

Description

Technical Field

[0001] The invention relates to a method for tracking at least one object using at least one detection device, the detection device being in particular a lidar system, in particular a detection device of a vehicle, wherein a plurality of target data are determined by using electromagnetic scanning signals, wherein each target data comprises at least one position value, which characterizes the position of the object target of the at least one object in at least one predetermined reference system.

[0002] Furthermore, the present invention relates to a detection device, in particular a lidar system, in particular a detection device for a vehicle, comprising a device for sending an electromagnetic scanning signal, a device for receiving a reflected electromagnetic scanning signal and a device for determining target data of an object target of an object based on the electromagnetic scanning signal.

[0003] In addition, the present invention relates to a vehicle having at least one detection device, in particular at least one lidar system, wherein the at least one detection device comprises a device for sending electromagnetic scanning signals, a device for receiving reflected electromagnetic scanning signals and a device for determining target data of an object target of an object based on the electromagnetic scanning signals. Background Art

[0004] A method for tracking an extended object with at least one detection device and a detection device are known from DE 10 2019 131 334 A1. In the method, an object is detected during continuous measuring cycles and at least one tracking state is determined from at least a portion of the detection of the object, which characterizes the position and movement of at least one object. During the measuring cycle, at least one transmission signal is emitted into a monitoring area of ​​at least one detection device, and at least one reception signal is received, which originates from at least one transmission signal reflected at at least one reflection point of at least one object. At least one detection of at least one reflection point is determined from the at least one reception signal. Advantageously, the detection device can include at least one radar sensor and / or at least one lidar sensor. Advantageously, the detection device can be used in a vehicle, in particular in a motor vehicle.

[0005] It is an object of the invention to provide a method, a detection device and a vehicle of the type mentioned above, with which the tracking of objects can be improved. Summary of the invention

[0006] The object of the present invention is achieved by such a method.

[0007] determining at least one boundary data characterizing a shape of at least one object based on at least a portion of the target data,

[0008] tracking at least one object by determining a progression of at least one boundary data,

[0009] wherein at least a portion of the target data is grouped into clusters in order to determine at least one boundary data,

[0010] determining at least one weighted value of the target data of at least one cluster from the target data of the corresponding cluster,

[0011] The at least one boundary data is determined from weighted target data, wherein each weighted target data is determined from corresponding target data weighted by at least one weight value of a cluster to which the target data belongs.

[0012] According to the invention, at least one boundary data is determined from at least a portion of the target data. The boundary data is used as a simplified model of at least one object, in particular a box, on which the scanning signal is reflected. The boundary data simplifies the tracking and segmentation of the at least one object compared to tracking and segmenting the entire target data set. The boundary data can characterize properties of the at least one object to be tracked, such as speed and direction. With the invention, the negative effects of irregular measurements can be compensated, in particular measurements of shape, which deviate from the simplified model of at least one object, for example from a box shape model. Such irregular measurements can make the tracking function unstable. With the invention, the measurements are corrected before matching the model. Therefore, further sterilization is unnecessary.

[0013] With the present invention, irregular shapes of the model produced by a part of a dynamic object can be corrected. For example, for an object in the form of a vehicle, irregular shapes can be produced by protruding parts such as side mirrors, tow bars, etc. With the present invention, low weights are applied to these measured parts according to the match with the model (e.g., a box model). The low weights reduce the influence of these measured parts on the match compared to the main part of at least one object.

[0014] In the sense of the invention, an object target is an area or a reflection point of an object from which a scanning signal can be reflected. An object can have one or more such object targets.

[0015] One advantage of the method according to the invention is that non-optimal matching is prevented, which may be difficult to compensate for later in the object tracking algorithm. With the method according to the invention, the measurements are directly processed and corrected to have a good match and further tracking. This is more robust and handles detection equipment, in particular LiDAR systems, with high sensitivity for measuring small shape irregularities (i.e. side mirrors, tow bars, etc.).

[0016] The method according to the invention can be used together with a detection device in the form of a lidar system. The detection device according to the invention can be a lidar system. With a lidar system, the position of an object relative to a predetermined reference system, in particular the direction and distance and the speed can be obtained. A lidar system can use electromagnetic scanning signals in the form of light signals, in particular laser signals. The light signals can be reflected by the objects. However, the method according to the invention can also be used together with a detection device in the form of a radar system. The detection device according to the invention can also be a radar system. With a radar system, the position of an object relative to a predetermined reference system, in particular the direction and distance and the speed can be obtained. A radar system can use electromagnetic scanning signals in the form of radar signals. The radar signals can be reflected by the objects.

[0017] Advantageously, the predetermined reference system may be a reference system of the at least one detection device and / or a reference system of a vehicle carrying the at least one detection device. In this way, the position of the object target may be clearly defined.

[0018] Advantageously, at least one reference system can be a coordinate system, in particular a polar coordinate system, a spherical coordinate system or a Cartesian coordinate system. The coordinate origin of the coordinate system can be associated with at least one detection device and / or the vehicle. At least one dimension of the coordinate system, in particular at least one axis, can be associated with at least one vehicle axis, such as a longitudinal vehicle axis, in particular a vehicle axis parallel thereto.

[0019] The method and the detection device according to the invention can be used for vehicles, in particular motor vehicles. Advantageously, the method and the detection device according to the invention can be used for land vehicles, in particular passenger cars, trucks, buses, motorcycles, etc., aircraft and / or ships. The method according to the invention can also be used for detection devices of vehicles that can be operated autonomously or partially autonomously. However, the method and the detection device according to the invention are not limited to vehicles. They can also be used for fixed operations, robots and / or machines, in particular construction or transportation machines, such as cranes, excavators, etc. The detection device can advantageously be connected to at least one electronic control device of a vehicle or machine, in particular a driver assistance system, or be a part thereof. In this way, automatic or partially automatic operation of a vehicle or machine can be achieved.

[0020] The detection device can be used to detect stationary or moving objects, in particular vehicles, people, animals, obstacles, road irregularities, in particular potholes or stones, road restrictions, open spaces, in particular parking spaces, precipitation, etc.

[0021] According to an advantageous embodiment, at least one weighting value of at least one cluster of target data, in particular contour target data, can be determined.

[0022] According to a plurality of target data in at least one cluster, in particular object target data

[0023] and / or

[0024] According to an extension of a range of comparable position values, in particular height values, of at least one cluster of target data, in particular object target data, in at least one dimension, wherein the extension of the range of position values ​​characterizes the spatial extent, in particular the height extension, of the distribution of the object targets to which the at least one cluster of target data, in particular object target data, belongs,

[0025] and / or

[0026] Based on a correlation of position values ​​of at least a portion of the target data, in particular object target data, of at least one cluster relative to one another, wherein the correlation characterizes an alignment of the object targets to which the target data of at least one cluster belong relative to one another,

[0027] and / or

[0028] Based on a cluster position value characterizing the position of the cluster relative to at least a portion of the other clusters. In this way, different weighting criteria can be applied to the target data representing the contour of the detected object. The criterion or combination of criteria to be used can be selected to suit the situation and / or the detected object.

[0029] Advantageously, at least one weighting value may be determined from a plurality of target data in at least one cluster. The smaller the number of target data in a cluster, the smaller the influence of the target data of the cluster on the boundary data.

[0030] Advantageously, at least one weighting value can be determined from the extension of the range of comparable position values ​​of at least one cluster of target data in at least one dimension. In particular, the position value can be a height value. The extension of the range of position values ​​can characterize the spatial extent of the distribution of the object data to which at least one cluster belongs, in particular the height extension. The smaller the spatial area spanned by the clustered object data, the smaller the influence of the corresponding target data of the corresponding cluster on the boundary data.

[0031] Advantageously, at least one weighting value can be determined from a correlation of position values ​​of at least a portion of target data of at least one cluster relative to one another, wherein the correlation characterizes the alignment of the object targets to which the target data of at least one cluster belong relative to one another. This allows target data deviating from a series of other target data representing the contour shape of the detected object to have a smaller influence on the boundary data.

[0032] Advantageously, at least one weighting value may be determined from a cluster position value, the cluster position value characterizing the position of the cluster relative to at least a portion of other clusters. This allows object objects that deviate from the object shape to be weighted less.

[0033] According to another advantageous embodiment, at least one weighting value of at least one cluster of target data, in particular contour target data, can be determined by statistical methods, in particular, at least one weighting value can be determined from a mean value and / or a variance. In this way, a weighting criterion can be determined in an efficient manner. The mean value and the variance can be used to infer whether the target data, in particular contour target data, of at least one cluster belongs to an object target on the main part of the detected object. Therefore, target data, in particular control target data, from an object target that does not belong to the main part of the detected object can be downweighted, in particular can be ignored.

[0034] According to another advantageous embodiment for determining at least one boundary data, only those target data, in particular contour target data, can be considered whose weighted value lies within a predetermined value range, in particular above a predetermined threshold value or below a predetermined threshold value. Thus, the influence of target data, in particular contour target data, of object targets that do not belong to the main part of the detected object on the boundary data can be minimized.

[0035] According to another advantageous embodiment

[0036] A plurality of object target data may be determined by using the electromagnetic scanning signal, each object target data representing an object target of the at least one object,

[0037] At least one contour object data characterizing a contour of at least one object may be determined from at least a portion of the object object data.

[0038] And at least one boundary data can be determined, in particular calculated, from at least one contour target data. The contour of at least one object can be approximated by determining the contour target data from a plurality of object target data. The object target data is data measured by using electromagnetic scanning signals. The boundary data can be determined from the contour target data approximating the contour of at least one object. In this way, the negative impact of irregular measurements can be further reduced.

[0039] According to another advantageous embodiment, a plurality of object target data can be determined by using the electromagnetic scanning signal, each object target data comprising at least one position value characterizing the three-dimensional position of an object target, and in order to determine at least one contour target data, the dimension of at least a part of the object target data can be reduced by one. In this way, object target data obtained from object targets of different levels of dimensionality to be reduced, in particular of different heights, can be combined into corresponding contour target data of one level, in particular of the same height.

[0040] According to another advantageous embodiment, at least a portion of at least one boundary data may be used to characterize at least one tracking region, in particular a tracking box, which is used to track at least one detected object.

[0041] and / or

[0042] At least one boundary data may include at least one position value and / or at least one direction value and / or at least one size value, wherein the at least one position value characterizes the position of the tracking area, in particular the tracking frame for tracking the at least one detected object, relative to at least one reference system, the at least one direction value characterizes the direction of the tracking area in at least one reference system, and the at least one size value characterizes the size of the tracking area. By means of the tracking area, an object target of interest may be identified. In this way, the movement of the object may be tracked more effectively. The position value may be used to characterize the position of the tracking area near the object target of interest. The orientation value may be used to characterize the orientation of the tracking area. The size value may be used to characterize the size of the tracking area.

[0043] Advantageously, at least one boundary data may comprise coordinates of a reference point of the tracking area, in particular the tracking box, in a reference system of the lidar system and / or of the vehicle with the lidar system, in particular Cartesian coordinates, polar coordinates or spherical coordinates. The reference point of the tracking area may be the center or an edge of the tracking area, for example. Furthermore, at least one boundary data may comprise at least one orientation value, in particular a yaw angle, which characterizes the orientation of the tracking area in a reference system, in particular a reference system of the lidar system and / or of the vehicle with the lidar system. Furthermore, at least one boundary data may comprise at least one size value, in particular a length, a width and a height, which characterizes the size of the tracking area.

[0044] Advantageously, at least one boundary data may be or include a tuple (x', y', z', α, l, w, h), where x', y' and z' are coordinates (Cartesian or spherical) of the position value, α is the yaw angle, l, w and h are dimension values ​​of the length, width and height of the tracking area. In the case of a two-dimensional tracking area, only two position values ​​and two dimension values ​​are required.

[0045] According to a further advantageous embodiment, at least one target data can be determined, which comprises at least one position value, in particular a direction value and / or a distance value, which characterizes the position of the target object in at least one reference system, and / or at least one intensity value, which characterizes the intensity of the electromagnetic scanning signal reflected at the target object. In this way, the object target can be assigned.

[0046] Advantageously, at least one target data may include at least one position value, in particular a direction value and / or a distance value. In this way, the position of the respective target object can be identified. At least one direction value can be implemented as an angle, in particular an azimuth and / or an elevation angle. In this way, the position of the object object can be identified in a spherical coordinate system or a polar coordinate system. At least one distance value can be implemented as a distance of the object object to the coordinate origin of at least one reference system.

[0047] Alternatively or alternatively, the at least one target datum may comprise at least one position value in the form of Cartesian coordinates of a Cartesian reference system, in particular a reference system of a lidar system and / or a reference system of a vehicle with an igniter system.

[0048] Advantageously, the target data may have the form of a tuple (p1, p2, p3, I), where p1, p2 and p3 are position values ​​in Cartesian or spherical coordinates and I is an intensity value. The intensity value characterizes the intensity of the received electromagnetic scanning signal. In the case of a two-dimensional reference system, only two position values ​​are required, in particular as an angle and a distance as polar coordinates or two Cartesian coordinates.

[0049] According to another advantageous embodiment, at least one target data can be determined, which target data includes at least one position value, which characterizes the two-dimensional position of the object target in at least one two-dimensional reference system, or the target data includes at least one position value, which characterizes the three-dimensional position of the object target in at least one three-dimensional reference system. In a two-dimensional reference system, the orientation of the target object in one dimension, in particular in azimuth, can be specified. In a three-dimensional reference system, the orientation of the target object in two dimensions, in particular in azimuth and elevation, can be specified.

[0050] Furthermore, the object of the invention is achieved by a testing device in that the testing device comprises means for carrying out the method according to the invention.

[0051] The detection device according to the present invention can be used to effectively track the motion of dynamic objects.

[0052] The apparatus for executing the method according to the present invention may be implemented by software and / or hardware.

[0053] Furthermore, the object of the invention is achieved by a vehicle in that the vehicle comprises a device for carrying out the method according to the invention.

[0054] According to the invention, a vehicle comprises at least one detection device and means for carrying out the method according to the invention. At least a part of the means for carrying out the method can be realized as part of at least one detection device. The detection device itself is part of the vehicle. Therefore, a part of the detection device is also part of the vehicle.

[0055] Additionally or alternatively, at least a part of the device for carrying out the method according to the invention can be realized separately from the at least one detection device, for example by means of a control device of the vehicle.

[0056] Advantageously, the vehicle may comprise at least one driver assistance system. With the driver assistance system, the information obtained from the at least one detection device may be used for autonomous or at least partially autonomous operation of the vehicle.

[0057] Advantageously, the at least one detection device can be connected to at least one driver assistance system. In this way, information acquired by means of the at least one detection device can be transmitted to the at least one driver assistance system.

[0058] Otherwise, the features and advantages shown in conjunction with the method according to the invention, the detection device according to the invention and the vehicle according to the invention and their respective advantageous configurations apply to each other after necessary modifications, and vice versa. Of course, the individual features and advantages can be combined with each other, thereby producing other beneficial effects that exceed the sum of the individual effects. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] The present invention and the above and other objects and advantages will be best understood from the following detailed description of the embodiments, but the present invention is not limited to the embodiments shown schematically.

[0060] Figure 1 is a top view of a vehicle with two lidar systems and a driver assistance system in a driving situation and a car with a trailer;

[0061] Figure 2 yes Figure 1 A side view of the vehicle;

[0062] Figure 3 is Figure 1 There is no trailer in the field of view of one of the vehicle's lidar systems. Figure 1 A side view of a car;

[0063] Figure 4 It is in the top view Figure 1 One of the vehicle's lidar systems obtains Figure 1 and Figure 3 Point cloud of car detection;

[0064] Figure 5 The top view shows the Figure 4 The weighted points representing the weighted contour target data determined in the point cloud;

[0065] Figure 6 is a tracking frame obtained using the method according to the first embodiment Figure 5 The weighted points of

[0066] Figure 7 Figure 4 The point cloud in Figure 6 Tracking box in

[0067] Figure 8 is a tracking frame obtained using the method according to the second embodiment Figure 5 The weighted points of

[0068] Fig. 9 yes Figure 4 Point cloud and Figure 8 Tracking box in ;

[0069] Fig.10 is obtained using a method known in the prior art with a tracking box from Figure 4 Point cloud.

[0070] In the accompanying drawings, the same or similar elements are represented by the same reference numerals. The accompanying drawings are merely schematic representations and are not intended to describe specific parameters of the present invention. In addition, the accompanying drawings are intended to describe only typical embodiments of the present invention and are therefore not to be considered as limiting the scope of the present invention. DETAILED DESCRIPTION

[0071] exist Figure 1 In FIG. 1 , a vehicle 10 in the form of a passenger car is shown in a top view, which in the driving state has two objects 18 in the form of cars with trailers. Figure 2 The vehicle 10 is shown in a side view with an implicit object 18 .

[0072] The vehicle 10 comprises two detection systems in the form of a driver assistance system 12 and a lidar system 14. The lidar system 14 is functionally connected to the driver assistance system 12 so that information acquired by the lidar system 14 on a monitoring area 16 in the direction of travel in front of and diagonally in front of the vehicle 10 can be transmitted to the driver assistance system 12. The driver assistance system 12 can be used to autonomously or semi-autonomously perform functions of the vehicle 10, such as driving functions.

[0073] For example, the lidar system 14 is arranged in the headlight of the vehicle 10 and is directed into the monitoring area 16. The two lidar systems 14 are located on opposite sides of the vehicle 10. The lidar system 14 can also be arranged at other locations of the vehicle 10, also in different directions. The lidar system 14 can be used to detect objects 18 in the monitoring area 16.

[0074] The object 18 may be a stationary or moving object, such as a vehicle, a person, an animal, a plant, an obstacle, a road irregularity (such as a pothole or a rock), a road boundary, a traffic sign, an open space (such as a parking space), precipitation, etc.

[0075] The laser radar systems 14 are identical in their design and mode of operation. The laser radar systems 14 and their functions are described below using one of the laser radar systems 14 as an example.

[0076] To detect the object 18, an electromagnetic scanning signal 20 in the form of laser pulses is emitted by the lidar system 14 into the monitoring area 16. The scanning signal 20 is received by the lidar system 14, which is reflected at an object target 22 of the object 18 in the direction of the lidar system 14. Object information such as the distance D, radial velocity and direction of the detected object target 22 relative to a lidar reference system 24 of the lidar system 14 and a vehicle reference system 26 of the vehicle 10 can be determined from the received scanning signal 20. In addition, the movement of the object 18 can be determined.

[0077] In the sense of the present invention, an object target 22 is a region or a reflection point of the object 18 from which the scanning signal 20 can be reflected. The object 18 can have one or more such object targets 22 .

[0078] For example, the LiDAR reference system 24 is a spherical coordinate system whose coordinates are azimuth angle φ, elevation angle θ, and distance D. The vehicle reference system 26 is, for example, a Cartesian coordinate system whose coordinates are designated as x, y, and z. The x direction extends in the longitudinal direction of the vehicle 10. The y direction is parallel to the lateral axis of the vehicle 10. The z direction is perpendicular to the x direction and is perpendicular to the y direction and upward.

[0079] During operation of the vehicle 10, the driver assistance system 12 uses object information, such as the distance, speed, direction, and direction of motion of the object 18 relative to the vehicle reference system 26, for autonomous or semi-autonomous operation of the vehicle 10. Using the lidar system 14, objects 18 that may be extended or elongated (e.g., trailers) may be tracked.

[0080] A method of tracking an object 18 using one of the lidar systems 14 is described below.

[0081] Using the transmitting device of the LiDAR system 14, scanning signals 20 are sent in the monitoring area 16. Scanning signals 20 that hit the object target 22 of the object 18 (for example, a car with a trailer) are reflected. Those scanning signals 20 reflected in the direction of the LiDAR system 14 are received by the receiving device of the LiDAR system 14.

[0082] In the sense of the invention, an object target 22 is an area or reflection point of the object 18 from which the scanning signal 20 can be reflected. The object 18 can have one or more such object targets 22. For the sake of clarity, only a few object targets 22 of the object 18 are shown in the figure as examples.

[0083] With the aid of the LiDAR system 14, the electromagnetic scanning signal 20 is converted into object target data (φ, θ, D, I). Each object target data (φ, θ, D, I) includes three position values, for example, representing the position of an object target 22 of the object 18 in the LiDAR reference system 24. The position values ​​are an azimuth value φ, an elevation value θ, and a distance value d. The azimuth value φ and the elevation value θ each represent the corresponding direction of the object target 22 relative to the coordinate origin of the LiDAR reference system 24. The distance value D represents the distance of the object target 22 from the coordinate origin of the LiDAR reference system 24. In addition, each object target data (φ, θ, D, I) includes an intensity value I representing the intensity of the received scanning signal 20.

[0084] The scanning signal 20 is used to scan the monitoring area 16 , thereby determining a plurality of object target data (φ, θ, D, I), each object target data representing an object target 22 . Figure 3 The object 18 (car) is shown in a side view in the field of view of the LiDAR system 14 , namely a horizontal field of view 38 and a vertical field of view 42 .

[0085] Thereafter, a set of contour target data (φ*, D*, I*) characterizing the contour of an object 18 (e.g., a car) is determined, for example, by calculation from the object target data (φ, θ, D, I). To this end, the object target data (φ, θ, D, I) obtained from object targets 22 at different vertical heights are combined into a set of contour target data (φ*, D*, I*). The vertical height is characterized by the elevation angle θ. That is, compared with the dimension of the object target data (φ, θ, D, I), the dimension of the contour target data (φ*, D*, I*) is reduced by one. Each contour target data (φ*, D*, I*) includes a contour azimuth φ*, a contour distance D*, and a contour intensity I*.

[0086] exist Figure 4 , a point cloud with object target data points 30 is shown in a two-dimensional view, each object target data point 30 visualizing object target data (φ, θ, D, I) of the object target 22. For clarity, only a few object target data points 30 are labeled. A contour target data set (φ*, D*, I*) representing the contour of the object 18 (i.e., the car) is shown with a solid line 28. The line 28 is composed of many contour data points, each contour data point representing contour target data (φ*, D*, I*). For easier orientation, a top view of the object 18 is shown with a dashed line.

[0087] Then, boundary data (x', y', z', α, l, w, h) are calculated from the contour object data set (φ*, D*, I*). The boundary data (x', y', z', α, l, w, h) are used to characterize the tracking area in the form of a tracking box 32. Figures 6 to 10 An exemplary tracking frame 32 is shown.

[0088] Exemplarily, the boundary data (x', y', z', α, l, w, h) has the form of a tuple. Each boundary data (x', y', z', α, l, w, h) includes three position values, namely the x'-y'-z'-coordinates of a reference point of the tracking frame 32 in the vehicle coordinate system 26, such as the center 34 of the tracking frame 32. In addition, each boundary data (x', y', z', α, l, w, h) includes an orientation value, namely the yaw angle α, which characterizes the orientation of the tracking frame 32 in the vehicle coordinate system 26. In addition, each boundary data (x', y', z', α, l, w, h) includes three size values, namely the length l, width w and height h of the tracking frame 32. The height h of the tracking frame 32 is not specified in the vehicle coordinate system 26. Figures 6 to 10 The two-dimensional representation is shown in FIG.

[0089] To determine boundary data (x', y', z', α, l, w, h), the contour object data set (φ*, D*, I*) is grouped into clusters 36. Thus, the horizontal field of view 38 of the lidar system 14 is divided into field of view sectors 40. Each field of view sector 40 extends over the entire vertical field of view 42 of the lidar system 14. Figure 3 , the horizontal field of view 38 and field of view sectors 40 of the LiDAR system 14 are depicted. Figure 4 A vertical field of view 42 and a field of view sector 40 are shown. Contour object data (φ*, D*, I*) generated from object targets 22 in the same field of view sector 40 are grouped in the same cluster 36.

[0090] For the contour target data (φ*, D*, I*) of each cluster 36 , a weighting value V is determined from the contour target data (φ*, D*, I*) of the corresponding cluster 36 . Figure 5 Weighted points 46 are shown, which show weighted contour object data (φ*, D*, I*, V) weighted with weight values ​​V of clusters 36 in a two-dimensional top view. For easier orientation, the top view of the vehicle 18 is shown in dashed lines. The diameters 44 of the weighted points 46 each correspond to the size of the weight value v.

[0091] To determine the weight value V of the contour object data (φ*, D*, I*) of a cluster 36, one of the following criteria or a combination of these criteria is used:

[0092] – The number of object target data in cluster 36 (φ, θ, D, I)

[0093] an extension of the range of height values ​​of the object data (φ, θ, D, I) of the cluster 36 , for example an elevation angle θ, characterizing a height extension 48 of the object 22 to which the object data (φ, θ, D, I) of the cluster 36 belong,

[0094] – Correlation of the position values ​​of the object target data (φ, θ, D, I) of the cluster 36 relative to each other, characterizing the alignment of the object targets 22 to which the object target data (φ, θ, D, I) of the cluster 36 belong relative to each other

[0095] – A cluster position value, which characterizes the position of a cluster 36 relative to other clusters 36 .

[0096] exist Figure 3 , the height extension 48 of the cluster 36 where the object target 22 of the side mirror 50 of the object 18 (car) is located, the height extension 48 of the cluster 36 where the object target 22 of the tow bar 52 of the object 18 is located, and the height extension 48 of the cluster 36 where the object target 22 of the driver's door 54 is located are marked as an example.

[0097] The listed criteria can be taken into account by using statistical quantities such as mean and variance. Therefore, the weight value V of the profile target data (φ*, D*, I*) of each cluster 36 is determined by statistical methods.

[0098] The boundary data (x', y', z', α, l, w, h) are determined from the weighted contour target data (φ*, D*, I*, V). Each of the weighted contour target data (φ*, D*, I*, V) is determined from the corresponding contour target data (φ*, D*, I*) weighted by the weight value V of the cluster 36 to which the target data belongs. Weighting the contour target data (φ*, D*, I*) reduces the influence of those target data originating from the irregular target 22 (such as the mirror 50 or the traction rod 52) on the boundary data (x', y', z', α, l, w, h).

[0099] exist Figure 6 The top view of FIG. 3 depicts the tracking frame 32 and the Figure 5 The tracking box 32 has the boundary data (x', y', z', α, l, w, h) of the object 18 (ie, the car). Figure 7 Shows Figure 4 The tracking box 32 and the object data points 30 in FIG.

[0100] The object 18, ie the car, is tracked with the lidar system 14 by following the tracking frame 32. Following the tracking frame 32 means determining the development of the boundary data (x', y', z', α, l, w, h).

[0101] Optionally, only those weighted contour target data (φ*, D*, I*, V) can be considered for determining the boundary data (x', y', z', α, l, w, h) whose weighted value V lies within a predetermined value range, for example above a predetermined threshold. This reduces the amount of data to be processed. Figure 8 The corresponding tracking box 32 of the object 18 (ie, the car) and Figure 5 The weighted points 46 in . The weighted points 46 having a weight value V below the threshold value are not considered for determination of the tracking frame 32 and each is crossed out. Fig. 9 Shows Figure 4 The corresponding tracking box 32 and object data point 30 in FIG.

[0102] Just for comparison, Fig.10 The tracking frame 32 obtained by the method known in the prior art is shown without weighting by the weight value V, and from Figure 4 The object data points are 30.

Claims

1. A method for tracking at least one object (18) using at least one detection device (14), in particular a laser radar system, in particular a detection device of a vehicle (10), wherein a plurality of target data (φ, θ, D, I) are determined by using an electromagnetic scanning signal (20), wherein each target data (φ, θ, D, I) comprises at least one position value (φ, θ, D), wherein the position value characterizes the position of an object target (22) of the at least one object (18) in at least one predetermined reference system (24, 26). Features determining at least one boundary data (x', y', z', α, l, w, h) based on at least a portion of the target data (φ, θ, D, I), which characterizes the shape of at least one object (18), tracking at least one object (18) by determining the development of at least one boundary data (x', y', z', α, l, w, h), wherein at least a portion of the target data (φ, θ, D, I) is grouped into clusters (36) in order to determine at least one boundary data (x', y', z', α, l, w, h), determining at least one weighting value (V) of the target data (φ, θ, D, I) of at least one cluster (36) from the target data (φ, θ, D, I) of the corresponding cluster (36), At least one boundary data (x', y', z', α, l, w, h) is determined by weighted target data (φ*, D*, I*, V*), wherein each weighted target data (φ*, D*, I*, V*) is determined by the corresponding target data (φ, θ, D, I) weighted by at least one weight value (V) of the cluster (36) to which the target data (φ, θ, D, I) belongs.

2. The method according to claim 1, characterized in that Determining at least one weighting value (V) of target data, in particular contour target data (φ*, D*, V*) of at least one cluster (36) According to a plurality of target data in at least one cluster (36), in particular object target data (φ, θ, D, I) and / or According to the extension of the range of comparable position values ​​(θ), in particular height values, of target data of at least one cluster (36), in particular object target data (φ, θ, D, I) in at least one dimension, wherein the extension of the range of position values ​​(θ) characterizes the spatial extent, in particular height extension (48), of the distribution of the object targets (22) to which the target data of at least one cluster (36), in particular object target data (φ, θ, D, I) belong, and / or Based on the position values ​​(φ, θ, D) of at least a part of the target data of at least one cluster (36), in particular the correlation of the object target data (φ, θ, D, I) relative to each other, wherein the correlation characterizes the alignment of the object targets (22) to which the target data (φ, θ, D, I) of at least one cluster (36) belong relative to each other, and / or Based on a cluster position value characterizing the position of the cluster (36) relative to at least a portion of the other clusters (36).

3. The method according to claim 1 or 2, characterized in that: At least one weighted value (V) of target data of at least one cluster (36), in particular contour target data (φ*, D*, V*), is determined by a statistical method, in particular, the at least one weighted value (V) is determined by a mean value and / or a variance.

4. The method according to any one of the preceding claims, characterized in that In order to determine the at least one boundary data (x', y', z', α, l, w, h), only those target data, in particular contour target data (φ*, D*, V*), are considered whose weighted value (V) is within a predetermined value range, in particular above a predetermined threshold or below a predetermined threshold.

5. The method according to any one of the preceding claims, characterized in that By using the electromagnetic scanning signal (20), a plurality of object target data (φ, θ, D, I) are determined, each object target data characterizing an object target (22) of at least one object (18), Determining at least one contour target data (φ*, D*, I*) characterizing a contour of at least one object (18) from at least a portion of the object target data (φ, θ, D, I) And at least one boundary data (x', y', z', α, l, w, h) is determined, in particular calculated, from at least one contour target data (φ*, D*, I*).

6. The method according to claim 5, characterized in that A plurality of object target data (φ, θ, D, I) are determined by using an electromagnetic scanning signal (20), each object target data comprising at least one position value (φ, θ, D) representing a three-dimensional position of an object target (22), and in order to determine at least one contour target data (φ*, D*, I), the dimension of at least a portion of the object target data (φ, θ, D, I) is reduced by one.

7. The method according to any one of the preceding claims, characterized in that At least a portion of the at least one boundary data (x', y', z', α, l, w, h) is used to characterize at least one tracking area (32), in particular a tracking box, which is used to track the at least one detected object (18) and / or At least one boundary data (x', y', z', α, l, w, h) comprises at least one position value (x', y', z') characterizing the position of a tracking area (32), in particular a tracking frame for tracking at least one detected object (18), relative to at least one reference system (24, 26), and / or at least one orientation value (α) characterizing the orientation of the tracking area (32) in at least one reference system (24, 26), and / or at least one size value characterizing the size of the tracking area (32).

8. The method according to any one of the preceding claims, characterized in that At least one target data (φ, θ, D, I; φ*, D*, I*; φ*, D*, I*, V*) is determined, which includes at least one position value (φ, θ, D, φ*, D*), in particular a direction value (φ, θ, φ*) and / or a distance value (D, D*), which characterizes the position of the object target (22) in at least one reference system (24, 26), and / or at least one intensity value (I, I*), which characterizes the intensity of an electromagnetic scanning signal (20) reflected at the object target (22).

9. The method according to any one of the preceding claims, characterized in that At least one target data (φ, θ, D, I; φ*, D*, I*; φ*, D*, I*, V*) is determined, which includes at least one position value, which represents the two-dimensional position of the object target (22) in at least one two-dimensional reference system, or it includes at least one position value (φ, θ, φ*, D*), which represents the three-dimensional position of the object target (22) in at least one three-dimensional reference system (24, 26).

10. A detection device (14), in particular a lidar system, in particular a detection device for a vehicle (10), comprising means for sending an electromagnetic scanning signal (20), means for receiving a reflected electromagnetic scanning signal (20) and means for determining target data of an object target (22) of an object (18) from the electromagnetic scanning signal (20), characterized in that The detection device (14) comprises means for carrying out the method according to one of claims 1 to 9.

11. A vehicle (10) having at least one detection device (14), in particular at least one lidar system, wherein the at least one detection device (14) comprises a device for transmitting an electromagnetic scanning signal (20), a device for receiving a reflected electromagnetic scanning signal (20) and a device for determining target data of an object target (22) of an object (18) from the electromagnetic scanning signal (20), characterized in that The vehicle (10) comprises means for carrying out the method according to one of claims 1 to 9.

Images