Method and system for testing dynamic target object in non-intervisibility environment

Abstract

The invention discloses a method and a system for testing a dynamic target object in a non-intervisibility environment. The method comprises the following steps: receiving perception information of all cooperative vehicles in a cooperative system on the dynamic target object; screening out the sensing information with the highest effectiveness from the sensing information of all cooperative vehicles; and combining the sensing information with the highest effectiveness with the sensing information of the main vehicle to obtain the state information of the dynamic target object relative to the main vehicle. The cooperative vehicle is used for providing the sensing information of the dynamic target object for the main vehicle, the main vehicle is assisted to obtain the state information of the dynamic target object, the problem that single vehicle sensing is shielded or cannot be sensed in a long distance in a non-intervisibility environment is solved through the multi-vehicle cooperative sensing technology, and therefore the accuracy and stability of environment sensing are improved, and the user experience is improved. The phenomena of false detection and missing detection are effectively reduced, and meanwhile, the robustness and safety of a traffic system are also improved.

Description

Technical Field

[0001] The present application relates to the field of automobile perception technology, and more specifically, to a method and system for testing dynamic targets in a non-line-of-sight environment. Background Art

[0002] In an open road traffic environment, especially in a non-line-of-sight environment (such as a curved road with buildings around it), the perception signal of the ego vehicle may be blocked by the surrounding environment, so the ego vehicle cannot obtain the information of traffic participants within the preset range; or in a non-simultaneous environment, traffic participants that may affect the driving safety of the ego vehicle are outside the perception range of the ego vehicle, so the ego vehicle cannot obtain their information. In the above situations, accurate perception information cannot be obtained, resulting in the inability to ensure the driving safety of the ego vehicle. Summary of the invention

[0003] The present application provides a testing method and system for dynamic targets in a non-line-of-sight environment, which utilizes cooperative vehicles to provide perception information of dynamic targets to a main vehicle, assisting the main vehicle in obtaining status information of dynamic targets, and utilizes multi-vehicle cooperative perception technology to solve the problem of single-vehicle perception being blocked or unable to perceive at a long distance in a non-line-of-sight environment, thereby improving the accuracy and stability of environmental perception, effectively reducing false detection and missed detection, and also improving the robustness and safety of the transportation system.

[0004] The present application provides a method for testing a dynamic target in a non-line-of-sight environment, comprising:

[0005] Receive the perception information of dynamic targets by all cooperative vehicles in the cooperative system;

[0006] Filter out the most effective perception information from all the perception information of the cooperative vehicles;

[0007] The most effective perception information is combined with the perception information of the host vehicle to obtain the state information of the dynamic target relative to the host vehicle.

[0008] Preferably, the most effective perception information is selected from the perception information of all cooperative vehicles, including:

[0009] The cooperative vehicle that is closest to the dynamic target or has the best perception angle is regarded as the best cooperative vehicle;

[0010] The perception information of the best cooperative vehicle on the dynamic target is taken as the most effective perception information.

[0011] Preferably, the method for testing dynamic targets in a non-line-of-sight environment further includes: using a true value device on the vehicle to perform external parameter tuning on its own perception device.

[0012] Preferably, the true value device on the vehicle is used to perform external parameter tuning on the perception device itself, specifically including:

[0013] Calibrate the external parameters of the laser radar in the true value device;

[0014] Project the perception images obtained by the laser radar and other perception devices on the vehicle onto the same plane;

[0015] Based on the calibrated external parameters of the LiDAR, fine-tune the external parameter values ​​of other sensing devices so that the sensing imaging of other sensing devices coincides with the sensing imaging of the LiDAR.

[0016] The extrinsic parameter values ​​of each other sensing device in the overlap state are taken as the tuned extrinsic parameter values.

[0017] Preferably, the method for testing a dynamic target in a non-line-of-sight environment further includes:

[0018] The optimized extrinsic parameter values ​​are used to calibrate the extrinsic parameters of the sensing device.

[0019] The present application also provides a dynamic target testing system in a non-line-of-sight environment, including a main vehicle, a dynamic target, and a cooperative system, wherein the cooperative system includes a plurality of cooperative vehicles, and the cooperative vehicles are communicatively connected with the main vehicle;

[0020] The main vehicle includes an information receiving module, a screening module and a relative status information obtaining module;

[0021] The receiving module is used to receive the perception information of all cooperative vehicles in the cooperative system on the dynamic target;

[0022] The screening module is used to screen out the most effective perception information from all the perception information of the cooperative vehicles;

[0023] The relative state information acquisition module is used to combine the most effective perception information with the perception information of the host vehicle to obtain the state information of the dynamic target relative to the host vehicle.

[0024] Preferably, the screening module includes a best vehicle determination module and a best information determination module;

[0025] The best vehicle determination module is used to select the cooperative vehicle with the shortest distance to the dynamic target or the best perception angle as the best cooperative vehicle;

[0026] The best information determination module is used to use the perception information of the best cooperative vehicle on the dynamic target as the most effective perception information.

[0027] Preferably, true value devices are provided on the host vehicle, cooperative vehicles and dynamic targets.

[0028] Preferably, the main vehicle, the cooperative vehicle and the dynamic target all include an external parameter tuning module, and the external parameter tuning module includes a first calibration module, a projection module, an optimization module and an external parameter determination module;

[0029] The first calibration module is used to calibrate the external parameters of the laser radar in the true value device;

[0030] The projection module is used to project the perception images obtained by the laser radar and other perception devices on the vehicle onto the same plane;

[0031] The optimization module is used to fine-tune the extrinsic parameters of other perception devices based on the calibrated extrinsic parameters of the lidar, so that the perception imaging of other perception devices coincides with the perception imaging of the lidar;

[0032] The external parameter determination module is used to use the external parameter value of each other sensing device in the overlap state as the tuned external parameter value.

[0033] Preferably, the main vehicle, cooperative vehicles and dynamic targets all use the optimized extrinsic parameter values ​​to calibrate the extrinsic parameters of the perception devices.

[0034] Other features and advantages of the present application will become apparent from the following detailed description of exemplary embodiments of the present application with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] The accompanying drawings, which are incorporated in and constitute a part of the specification, illustrate embodiments of the application and, together with the description, serve to explain the principles of the application.

[0036] Figure 1 A schematic diagram of a testing system for dynamic targets in a non-line-of-sight environment provided in this application;

[0037] Figure 2 A schematic diagram of a method for testing dynamic targets in a non-line-of-sight environment provided in this application;

[0038] Figure 3 A schematic diagram of the structure of the main vehicle provided for this application;

[0039] Figure 4 A schematic diagram of the structure of the external parameter tuning module provided for this application;

[0040] Figure 5 This is a schematic diagram of the structure of the external parameter calibration module provided in this application. DETAILED DESCRIPTION

[0041] Various exemplary embodiments of the present application will now be described in detail with reference to the accompanying drawings. It should be noted that unless otherwise specifically stated, the relative arrangement of components and steps, numerical expressions and numerical values ​​set forth in these embodiments do not limit the scope of the present application.

[0042] The following description of at least one exemplary embodiment is merely illustrative in nature and is in no way intended to limit the present application, its application, or uses.

[0043] Technologies, methods, and equipment known to ordinary technicians in the relevant art may not be discussed in detail, but where appropriate, the technologies, methods, and equipment should be considered part of the specification.

[0044] In all examples shown and discussed herein, any specific values ​​should be interpreted as merely exemplary and not limiting. Therefore, other examples of the exemplary embodiments may have different values.

[0045] The present application provides a method and system for testing dynamic targets in a non-line-of-sight environment, using cooperative vehicles to provide the main vehicle with perception information of dynamic targets, assisting the main vehicle in obtaining the status information of dynamic targets, and using multi-vehicle cooperative perception technology to solve the problem of single-vehicle perception being blocked or unable to perceive at a long distance in a non-line-of-sight environment, thereby improving the accuracy and stability of environmental perception, effectively reducing false detection and missed detection, and also improving the robustness and safety of the traffic system. In addition, the vehicle uses true value equipment to perform external parameter tuning and external parameter calibration on the perception equipment to ensure that the vehicle perception system can provide reliable and accurate situational understanding in various driving environments, thereby enhancing the safety and driving efficiency of the vehicle.

[0046] like Figure 1 As shown, the test system for dynamic targets in a non-line-of-sight environment provided by the present application includes a main vehicle, a dynamic target, and a cooperative system, wherein the cooperative system includes a plurality of cooperative vehicles, and the cooperative vehicles are communicatively connected with the main vehicle. In some embodiments, if the dynamic target has a communication function, the dynamic target is also communicatively connected with the cooperative vehicle and the main vehicle.

[0047] It is understandable that there are many traffic participants in a non-line-of-sight environment, and all traffic participants together form a traffic system, in which one vehicle can be used as the main vehicle, another traffic participant (such as a motor vehicle, non-motor vehicle or pedestrian) can be used as a dynamic target, and the other vehicles form a collaborative system. Vehicles communicate with each other through communication methods such as Internet of Vehicles or true value devices (including V-BOX, LiDAR, etc.).

[0048] exist Figure 1In the example shown, the non-line-of-sight environment is a road environment with a right-angle turn. Three vehicles equipped with a human-machine co-driving system (the main vehicle, the first cooperative vehicle, and the second cooperative vehicle) are parked on the side of the road. The main vehicle and the first cooperative vehicle are parked on different straight roads respectively. The second cooperative vehicle is parked at the turning intersection and keeps a certain distance from the first cooperative vehicle and the main vehicle. The dynamic target is at a certain distance in front of the first cooperative vehicle. The vehicles realize multi-vehicle cooperative perception by sharing perception data.

[0049] Based on the above-mentioned dynamic target testing system in a non-line-of-sight environment, the present application provides a dynamic target testing method in a non-line-of-sight environment. Figure 2 As shown, the test methods for dynamic targets in non-line-of-sight environments include:

[0050] S210: The host vehicle receives the perception information of the dynamic target object from all cooperative vehicles in the cooperative system.

[0051] Among them, the cooperative vehicle's perception information of dynamic targets includes position (lateral and longitudinal distance from the cooperative vehicle), speed (relative speed or actual speed), and direction of movement.

[0052] During the test, when the cooperative vehicle obtains the perception information of the dynamic target, combined with Figure 1 In the example shown, the first cooperative vehicle uses the static position as the initial position and the rear axle center of the vehicle as the reference point. The first cooperative vehicle senses and outputs the position of the dynamic target through the sensing device (camera, laser radar, etc.) carried by the vehicle to determine the lane and direction of the dynamic target. When the dynamic target is moving at a constant speed in the lane at a fixed speed, the first cooperative vehicle senses the speed of the dynamic target and the distance between the two vehicles through the sensing device carried by the vehicle.

[0053] If the distance between the first cooperative vehicle and the main vehicle is large, its perception information cannot be directly transmitted to the main vehicle, then the perception information is forwarded to the main vehicle through the second cooperative vehicle between the two. If the distance between the first cooperative vehicle and the main vehicle is small, its perception information can be directly transmitted to the main vehicle, then the perception information is directly transmitted to the main vehicle.

[0054] The method for acquiring perception information of the second cooperative vehicle is the same as that of the first cooperative vehicle.

[0055] It should be noted that each vehicle's perception device has a corresponding perception range. If a dynamic target is outside the perception range of a certain perception device or is blocked by objects in a non-line-of-sight environment, the vehicle cannot obtain complete perception information.

[0056] It can be understood that in a non-line-of-sight environment, there may be at least one cooperative vehicle that can obtain complete perception information, there may be cooperative vehicles that cannot obtain complete perception information, or there may be no cooperative vehicles that cannot obtain complete perception information, depending on the specific circumstances.

[0057] S220: Filter out the most effective perception information from the perception information of all cooperative vehicles.

[0058] As an embodiment, the most effective perception information is selected from the perception information of all cooperative vehicles, specifically including:

[0059] R1: The cooperative vehicle that is closest to the dynamic target or has the best perception angle is regarded as the best cooperative vehicle. The accuracy and precision of the perception information obtained by the cooperative vehicle that is closest to the dynamic target or has the best perception angle is relatively higher.

[0060] R2: The perception information of the best cooperative vehicle on the dynamic target is taken as the most effective perception information.

[0061] S230: Combining the most effective perception information with the perception information of the host vehicle to obtain state information of the dynamic target object relative to the host vehicle.

[0062] As an embodiment, the most effective perception information includes the lane position, vehicle speed, movement direction, lateral distance and longitudinal distance between the dynamic target and the best cooperative vehicle, etc. The perception information of the main vehicle itself includes the lane position, vehicle speed, movement direction, lateral distance and longitudinal distance between the main vehicle and the best cooperative vehicle, etc. Through these two sets of perception information, the position and movement speed of the dynamic target relative to the main vehicle can be obtained, and this information serves as the status information of the dynamic target relative to the main vehicle.

[0063] Preferably, in the present application, the method for testing dynamic targets in a non-line-of-sight environment further includes: using a true value device on the vehicle to perform external parameter tuning on its own perception device.

[0064] As an embodiment, the true value device on the vehicle is used to perform external parameter tuning on the perception device itself, specifically including:

[0065] P1: Calibrate the external parameters of the lidar in the true value device.

[0066] P2: Project the perception images obtained by the LiDAR and other perception devices on the vehicle onto the same plane. At this time, for the same target object, the perception images of other perception devices have a certain degree of overlap with the perception images of the LiDAR, which may or may not completely overlap.

[0067] P3: Based on the calibrated external parameters of the LiDAR, fine-tune the external parameter values ​​of other perception devices so that the perception imaging of other perception devices completely overlaps with the perception imaging of the LiDAR.

[0068] P4: The extrinsic parameter values ​​of each other sensing device in the overlap state are taken as the tuned extrinsic parameter values.

[0069] Based on the optimized external parameter values, the installation position, installation angle, etc. of the sensing device are adjusted so that its sensing accuracy meets the design indicators, thereby improving the sensing accuracy of the sensing device.

[0070] Preferably, the method for testing dynamic targets in a non-line-of-sight environment further includes: calibrating the external parameters of the sensing device using the tuned external parameter values.

[0071] As an embodiment, the external parameter calibration of the sensing device is performed using the tuned external parameter value, which specifically includes:

[0072] Q1: Correspond the dynamic perception information obtained in each perception device’s own perception coordinate system to the vehicle body coordinate system.

[0073] Specifically, through online dynamic calibration, the perception information (target point cloud position information in the perception coordinate system) obtained by each perception device is used as a reference to transform the perception coordinate system into a unified vehicle body coordinate system.

[0074] Q2: Based on the corresponding relationship, the transformation matrix between the perception coordinate system and the vehicle coordinate system is obtained, and the transformation matrix is ​​used as the external parameter calibration information of the perception device (including installation position, installation angle, etc.).

[0075] After the perception coordinate system is converted into a unified vehicle body coordinate system, the matrix transformation relationship between the perception coordinate system and the vehicle body coordinate system, that is, the conversion matrix, can be obtained.

[0076] The perception information of dynamic targets identified on the vehicle is compared with the perception information obtained by its own true value equipment to achieve external parameter tuning and calibration, ensuring that the vehicle's perception system can provide reliable and accurate situational understanding in various driving environments, provide accurate perception information for the host vehicle, and improve the safety and driving efficiency of the host vehicle.

[0077] Based on the above, in this application, Figure 3 As shown, the host vehicle includes an information receiving module 310 , a screening module 320 and a relative status information obtaining module 330 .

[0078] The receiving module 310 is used to receive the perception information of all cooperative vehicles in the cooperative system on the dynamic target object.

[0079] The screening module 320 is used to screen out the most effective perception information from the perception information of all cooperative vehicles.

[0080] The relative state information acquisition module 330 is used to combine the most effective perception information with the perception information of the host vehicle to obtain the state information of the dynamic target relative to the host vehicle.

[0081] Preferably, the screening module 320 includes an optimal vehicle determination module 3201 and an optimal information determination module 3202 .

[0082] The best vehicle determination module 3201 is used to select the cooperative vehicle that is closest to the dynamic target or has the best perception angle as the best cooperative vehicle.

[0083] The best information determination module 3202 is used to use the perception information of the best cooperative vehicle on the dynamic target as the most effective perception information.

[0084] Preferably, the main vehicle, cooperative vehicles and dynamic targets are equipped with true value devices (including V-BOX, laser radar, etc.), and the true value devices of the vehicles can communicate with each other.

[0085] Preferably, the main vehicle, the cooperative vehicle and the dynamic target all include an external parameter tuning module. Figure 4 As shown, the external parameter tuning module includes a first calibration module 410 , a projection module 420 , an optimization module 430 and an external parameter determination module 440 .

[0086] The first calibration module 410 is used to calibrate the external parameters of the laser radar in the true value device.

[0087] The projection module 420 is used to project the perception images obtained by the laser radar and other perception devices on the vehicle onto the same plane.

[0088] The optimization module 430 is used to fine-tune the extrinsic parameter values ​​of other sensing devices based on the calibrated extrinsic parameters of the laser radar, so that the sensing imaging of other sensing devices coincides with the sensing imaging of the laser radar.

[0089] The external parameter determination module 440 is used to use the external parameter value of each other sensing device in the overlap state as the tuned external parameter value.

[0090] Preferably, the main vehicle, the cooperative vehicle and the dynamic target object all use the tuned external parameter values ​​to calibrate the external parameters of the sensing devices, and thus, the main vehicle, the cooperative vehicle and the dynamic target object all include an external parameter calibration module.

[0091] Preferably, if Figure 5 As shown, the external parameter calibration module includes a corresponding module 510 and a conversion matrix acquisition module 520.

[0092] The corresponding module 510 is used to correspond the dynamic perception information obtained in the perception coordinate system of each perception device to the vehicle body coordinate system.

[0093] The conversion matrix acquisition module 520 is used to obtain the conversion matrix between the perception coordinate system and the vehicle body coordinate system according to the corresponding relationship, and use the conversion matrix as the external parameter calibration information of the perception device.

[0094] The present application can determine road boundaries in non-line-of-sight environments, detect and identify dynamic targets, locate the lane where dynamic targets are located, detect the position, movement direction and vehicle speed of dynamic targets, etc. through multi-vehicle collaborative perception, thereby improving the perception accuracy in non-line-of-sight environments.

[0095] Although some specific embodiments of the present application have been described in detail by way of example, it should be understood by those skilled in the art that the above examples are only for illustration, not for limiting the scope of the present application. It should be understood by those skilled in the art that the above embodiments may be modified without departing from the scope and spirit of the present application. The scope of the present application is defined by the appended claims.

Claims

1. A method for testing dynamic targets in a non-line-of-sight environment, characterized in that: include: Receive the perception information of dynamic targets by all cooperative vehicles in the cooperative system; Filter out the most effective perception information from all the perception information of the cooperative vehicles; The most effective perception information is combined with the perception information of the host vehicle to obtain the state information of the dynamic target relative to the host vehicle.

2. The method for testing dynamic targets in a non-line-of-sight environment according to claim 1, characterized in that: Filter out the most effective perception information from all cooperative vehicles, including: The cooperative vehicle that is closest to the dynamic target or has the best perception angle is selected as the best cooperative vehicle; The perception information of the best cooperative vehicle on the dynamic target object is used as the most effective perception information.

3. The method for testing dynamic targets in a non-line-of-sight environment according to claim 1, characterized in that: Also includes: Use the true value equipment on the vehicle to optimize the external parameters of its own perception equipment.

4. The method for testing dynamic targets in a non-line-of-sight environment according to claim 3, characterized in that: Use the true value equipment on the vehicle to tune the external parameters of its own perception equipment, including: Calibrate the external parameters of the laser radar in the true value device; Projecting the perception images obtained by the laser radar and other perception devices on the vehicle onto the same plane; Based on the calibrated external parameters of the LiDAR, fine-tune the external parameter values ​​of other sensing devices so that the sensing imaging of other sensing devices coincides with the sensing imaging of the LiDAR. The extrinsic parameter values ​​of each other sensing device in the overlap state are taken as the tuned extrinsic parameter values.

5. The method for testing dynamic targets in a non-line-of-sight environment according to claim 3, characterized in that: Also includes: The optimized extrinsic parameter values ​​are used to calibrate the extrinsic parameters of the sensing device.

6. A dynamic target testing system in a non-line-of-sight environment, characterized in that: It includes a main vehicle, a dynamic target and a cooperative system, wherein the cooperative system includes a plurality of cooperative vehicles, and the cooperative vehicles are communicatively connected with the main vehicle; The main vehicle includes an information receiving module, a screening module and a relative status information obtaining module; The receiving module is used to receive the perception information of the dynamic target by all the cooperative vehicles in the cooperative system; The screening module is used to screen out the most effective perception information from the perception information of all cooperative vehicles; The relative state information acquisition module is used to combine the most effective perception information with the perception information of the host vehicle to obtain the state information of the dynamic target relative to the host vehicle.

7. The dynamic target testing system in a non-line-of-sight environment according to claim 6, characterized in that: The screening module includes a best vehicle determination module and a best information determination module; The best vehicle determination module is used to select the cooperative vehicle with the shortest distance to the dynamic target or the best perception angle as the best cooperative vehicle; The optimal information determination module is used to use the perception information of the optimal cooperative vehicle on the dynamic target object as the most effective perception information.

8. The dynamic target testing system in a non-line-of-sight environment according to claim 6, characterized in that: The main vehicle, the cooperative vehicle and the dynamic target object are provided with true value devices.

9. The dynamic target testing system in a non-line-of-sight environment according to claim 8, characterized in that: The main vehicle, the cooperative vehicle and the dynamic target all include an external parameter tuning module, and the external parameter tuning module includes a first calibration module, a projection module, an optimization module and an external parameter determination module; The first calibration module is used to calibrate the external parameters of the laser radar in the true value device; The projection module is used to project the perception images obtained by the laser radar and other perception devices on the vehicle onto the same plane; The optimization module is used to fine-tune the extrinsic parameter values ​​of other sensing devices based on the calibrated extrinsic parameters of the laser radar, so that the sensing imaging of other sensing devices coincides with the sensing imaging of the laser radar; The external parameter determination module is used to use the external parameter value of each other sensing device in the overlap state as the tuned external parameter value.

10. The dynamic target testing system in a non-line-of-sight environment according to claim 9, characterized in that: The main vehicle, the cooperative vehicle and the dynamic target object all use the tuned external parameter values ​​to calibrate the external parameters of the perception equipment.

Images