Intelligent driving vehicle testing method and device
By loading virtual driving scenarios into intelligent driving vehicles and adjusting strategies based on pose information from actual scenarios, the problems of high testing costs and insufficient reliability of intelligent driving vehicles are solved, achieving low-cost and high-efficiency testing results.
Patent Information
- Application Number
- CN202410804171.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-06-21
- Publication Date
- 2024-11-22
- Estimated Expiration
- 2044-06-21
Smart Images

Figure CN119000109A_ABST
Abstract
Description
[Technical field]
[0001] The present application relates to the field of intelligent driving technology, and in particular to an intelligent driving vehicle testing method and device. [Background technology]
[0002] In recent years, with the rapid development of intelligent driving technology, how to verify and evaluate the driving safety of intelligent driving vehicles has become the key. Obviously, the safety and reliability testing of intelligent driving technology needs to be carried out over a long period of time, extensively and repeatedly, and its related influencing factors include scenarios, equipment, personnel and other external conditions. At present, the scenarios in the preset scenario library can generally be used to test the algorithm files to be tested without setting up actual driving scenarios, so as to reduce the testing cost.
[0003] However, this fully virtual testing method relies on the richness of the preset scenario library, which can only cover a limited number of scenarios, and the accuracy and practicality of the test results are limited. At the same time, establishing rich scenario data will also bring a lot of computing work, which is extremely costly. Moreover, the use of a fully virtual testing method cannot fully and efficiently reproduce the various problems that intelligent driving vehicles may face in actual scenarios.
[0004] In this regard, related technologies also provide some technical solutions that combine virtuality and reality. While controlling the physical vehicle in the actual scene, the actual scene is reproduced as an operable virtual scene to eliminate traditional driver operations and improve test safety. However, this method has very high standards for the construction of the actual scene, which not only brings high cost consumption, but also cannot cover a large number of complex and changing scenes due to the limitation of scene construction capabilities. The range of scenes covered by the test is limited, which also affects the reliability of the test results.
[0005] Therefore, how to balance the test cost and the reliability of test results in intelligent driving vehicle testing has become a technical problem that needs to be solved urgently. [Summary of the invention]
[0006] The embodiments of the present application provide a smart driving vehicle testing method and device, which aim to solve the technical problems of high testing cost and insufficient reliability of test results in smart driving vehicle testing methods in related technologies.
[0007] In a first aspect, an embodiment of the present application provides a method for testing an intelligent driving vehicle, comprising:
[0008] In response to the intelligent driving test instruction, loading a corresponding virtual driving scene for the intelligent driving vehicle in the actual driving scene;
[0009] Acquire first real-time position information of the intelligent driving vehicle in the actual driving scene;
[0010] Based on the first real-time position information, determining second real-time position information of the intelligent driving vehicle in the virtual driving scene;
[0011] Generate an automatic driving adjustment strategy based on the second real-time posture information, the real-time driving environment information provided by the virtual driving scene, and a preset intelligent driving algorithm;
[0012] The intelligent driving vehicle is controlled to execute the automatic driving adjustment strategy in the virtual driving scenario and the actual driving scenario respectively.
[0013] In one embodiment of the present application, optionally, determining second real-time posture information of the intelligent driving vehicle in the virtual driving scene based on the first real-time posture information includes:
[0014] Based on the posture offset information between the actual driving scene and the virtual driving scene and the first real-time posture information, second real-time posture information of the intelligent driving vehicle in the virtual driving scene is determined.
[0015] In one embodiment of the present application, optionally, when loading a corresponding virtual driving scene for an intelligent driving vehicle in an actual driving scene, the method further includes:
[0016] Acquire first initial position information of the intelligent driving vehicle in the actual driving scene and second initial position information of the intelligent driving vehicle in the virtual driving scene;
[0017] The difference between the first initial posture information and the second initial posture information is used to determine the posture offset information of the intelligent driving vehicle between the actual driving scene and the virtual driving scene.
[0018] In one embodiment of the present application, optionally, when loading a corresponding virtual driving scene for an intelligent driving vehicle in an actual driving scene, the method further includes:
[0019] Acquire first initial position information of the intelligent driving vehicle in the actual driving scene and second initial position information of the intelligent driving vehicle in the virtual driving scene;
[0020] Obtaining a difference between the first initial posture information and the second initial posture information;
[0021] Determining a first difference adjustment coefficient corresponding to the virtual driving scene based on the type of the virtual driving scene;
[0022] Based on the difference between the first initial posture information and the second initial posture information, and the first difference adjustment coefficient, the posture offset information of the intelligent driving vehicle between the actual driving scene and the virtual driving scene is determined.
[0023] In one embodiment of the present application, optionally, when loading a corresponding virtual driving scene for an intelligent driving vehicle in an actual driving scene, the method further includes:
[0024] Acquire first initial position information of the intelligent driving vehicle in the actual driving scene and second initial position information of the intelligent driving vehicle in the virtual driving scene;
[0025] Obtaining a difference between the first initial posture information and the second initial posture information;
[0026] Determining a scene difference between the actual driving scene and the virtual driving scene based on preset scene coefficients corresponding to the actual driving scene and the virtual driving scene respectively;
[0027] Based on the scene difference, determining a second difference adjustment coefficient;
[0028] Based on the difference between the first initial posture information and the second initial posture information, and the second difference adjustment coefficient, the posture offset information of the intelligent driving vehicle between the actual driving scene and the virtual driving scene is determined.
[0029] In one embodiment of the present application, optionally, the step of loading a corresponding virtual driving scene for an intelligent driving vehicle in an actual driving scene includes:
[0030] A virtual-reality combined simulation system is built for the intelligent driving vehicle, wherein the virtual-reality combined simulation system includes a host computer, a rapid prototyping machine, the intelligent driving vehicle, a combined navigation module and a communication module.
[0031] The host computer is used to load a virtual driving scene file reproduced by real vehicle environment data and display the virtual driving scene corresponding to the virtual driving scene file, wherein the virtual driving scene file includes environment information and movement association information of historical traffic participants in the virtual driving scene;
[0032] The combined navigation module is used to locate the intelligent driving vehicle;
[0033] The rapid prototyping machine is used to run the preset intelligent driving algorithm based on the information provided by the host computer, the intelligent driving vehicle and the integrated navigation module, and control each target actuator of the intelligent driving vehicle to execute the corresponding automatic driving adjustment strategy based on the running result;
[0034] The host computer, the rapid prototype machine, the intelligent driving vehicle and the combined navigation module are connected via the communication module.
[0035] In one embodiment of the present application, optionally, the controlling each target execution mechanism of the intelligent driving vehicle to execute a corresponding automatic driving adjustment strategy includes:
[0036] Sending control instructions required for the automatic driving adjustment strategy to the driving system, braking system and steering system of the intelligent driving vehicle;
[0037] Sending a warning instruction to the warning module of the intelligent driving vehicle.
[0038] In a second aspect, an embodiment of the present application provides an intelligent driving vehicle testing device, comprising:
[0039] A driving scenario configuration unit, configured to load a corresponding virtual driving scenario for an intelligent driving vehicle in an actual driving scenario in response to an intelligent driving test instruction;
[0040] A first real-time position and posture information acquisition unit, used to acquire first real-time position and posture information of the intelligent driving vehicle in the actual driving scene;
[0041] A second real-time posture information determining unit, configured to determine second real-time posture information of the intelligent driving vehicle in the virtual driving scene based on the first real-time posture information;
[0042] an automatic driving adjustment strategy generating unit, configured to generate an automatic driving adjustment strategy based on the second real-time posture information, the real-time driving environment information provided by the virtual driving scene, and a preset intelligent driving algorithm;
[0043] The automatic driving adjustment strategy execution unit is used to control the intelligent driving vehicle to execute the automatic driving adjustment strategy in the virtual driving scene and the actual driving scene respectively.
[0044] In one embodiment of the present application, optionally, the second real-time posture information determination unit is used to:
[0045] Based on the posture offset information between the actual driving scene and the virtual driving scene and the first real-time posture information, second real-time posture information of the intelligent driving vehicle in the virtual driving scene is determined.
[0046] In one embodiment of the present application, optionally, the device further includes:
[0047] An initial posture information acquisition unit, configured to acquire, when loading a corresponding virtual driving scene for an intelligent driving vehicle in an actual driving scene, first initial posture information of the intelligent driving vehicle in the actual driving scene and second initial posture information in the virtual driving scene;
[0048] The first determining unit is used to determine the posture offset information of the intelligent driving vehicle between the actual driving scene and the virtual driving scene based on the difference between the first initial posture information and the second initial posture information.
[0049] In one embodiment of the present application, optionally, the device further includes:
[0050] An initial posture information acquisition unit, configured to acquire, when loading a corresponding virtual driving scene for an intelligent driving vehicle in an actual driving scene, first initial posture information of the intelligent driving vehicle in the actual driving scene and second initial posture information in the virtual driving scene;
[0051] A difference acquisition unit, used to acquire a difference between the first initial posture information and the second initial posture information;
[0052] A first difference adjustment coefficient determining unit, configured to determine a first difference adjustment coefficient corresponding to the virtual driving scene based on the type of the virtual driving scene;
[0053] The second determination unit is used to determine the posture offset information of the intelligent driving vehicle between the actual driving scene and the virtual driving scene based on the difference between the first initial posture information and the second initial posture information, and the first difference adjustment coefficient.
[0054] In one embodiment of the present application, optionally, the device further includes:
[0055] An initial posture information acquisition unit, configured to acquire, when loading a corresponding virtual driving scene for an intelligent driving vehicle in an actual driving scene, first initial posture information of the intelligent driving vehicle in the actual driving scene and second initial posture information in the virtual driving scene;
[0056] A difference acquisition unit, used to acquire a difference between the first initial posture information and the second initial posture information;
[0057] a scene difference determination unit, configured to determine a scene difference between the actual driving scene and the virtual driving scene based on preset scene coefficients corresponding to the actual driving scene and the virtual driving scene respectively;
[0058] a second difference adjustment coefficient determining unit, configured to determine a second difference adjustment coefficient based on the scene difference;
[0059] The third determination unit is used to determine the posture offset information of the intelligent driving vehicle between the actual driving scene and the virtual driving scene based on the difference between the first initial posture information and the second initial posture information, and the second difference adjustment coefficient.
[0060] In one embodiment of the present application, optionally, the driving scene configuration unit includes:
[0061] A virtual-reality combined simulation system construction unit is used to build a virtual-reality combined simulation system for the intelligent driving vehicle, wherein the virtual-reality combined simulation system includes a host computer, a rapid prototyping machine, the intelligent driving vehicle, a combined navigation module and a communication module.
[0062] The host computer is used to load a virtual driving scene file reproduced by real vehicle environment data and display the virtual driving scene corresponding to the virtual driving scene file, wherein the virtual driving scene file includes environment information and movement association information of historical traffic participants in the virtual driving scene;
[0063] The combined navigation module is used to locate the intelligent driving vehicle;
[0064] The rapid prototyping machine is used to run the preset intelligent driving algorithm based on the information provided by the host computer, the intelligent driving vehicle and the integrated navigation module, and control each target actuator of the intelligent driving vehicle to execute the corresponding automatic driving adjustment strategy based on the running result;
[0065] The host computer, the rapid prototype machine, the intelligent driving vehicle and the combined navigation module are connected via the communication module.
[0066] In one embodiment of the present application, optionally, the autonomous driving adjustment strategy execution unit is used to:
[0067] The control instructions required for the automatic driving adjustment strategy are sent to the driving system, braking system and steering system of the intelligent driving vehicle; and the warning instructions are sent to the warning module of the intelligent driving vehicle.
[0068] In a third aspect, an embodiment of the present application provides a computer device, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are configured to execute the method described in the first aspect above.
[0069] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium storing computer-executable instructions, wherein the computer-executable instructions are used to execute the method described in the first aspect above.
[0070] The above technical scheme aims at the technical problems of high test cost and insufficient reliability of test results in the intelligent driving vehicle testing method in the related technology. On the basis of combining the virtual driving scene with the actual driving scene for remote testing, there is no need to replicate the actual driving scene consistent with the virtual driving scene, but only needs to control the intelligent driving vehicle in the ordinary test site to move according to the real-time driving environment information provided by the virtual driving scene, and the same result of controlling the intelligent driving vehicle to drive in the actual scene corresponding to the virtual driving scene can be obtained. In other words, there is no need to replicate complex scenes for intelligent driving vehicles, but to allow them to obtain the data of complex scenes in ordinary actual driving scenes, so as to achieve the effect of testing intelligent driving vehicles in complex scenes. At the same time, the present application also takes into account the difference between the virtual driving scene and the actual driving scene, and after correcting the first real-time posture information of the intelligent driving vehicle in the actual driving scene, the second real-time posture information of the intelligent driving vehicle in the virtual driving scene is obtained, so as to reduce the error caused by the data conversion between the virtual driving scene and the actual driving scene during the test, and improve the accuracy of the test results.
Brief Description of the Drawings
[0071] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0072] Figure 1 A flow chart of a method for testing an intelligent driving vehicle according to an embodiment of the present application is shown;
[0073] Figure 2 A schematic diagram of a virtual-reality combined simulation system according to an embodiment of the present application is shown;
[0074] Figure 3 A flowchart of obtaining posture deviation information according to an embodiment of the present application is shown;
[0075] Figure 4 A network topology diagram of an intelligent driving vehicle vehicle-in-the-loop test system based on real vehicle environment data injection according to an embodiment of the present application is shown;
[0076] Figure 5A network topology diagram of an intelligent driving vehicle whole vehicle in-the-loop test system based on real vehicle environment data injection according to another embodiment of the present application is shown;
[0077] Figure 6 A schematic diagram of a historical scene according to an embodiment of the present application is shown;
[0078] Figure 7 A schematic diagram showing a combination of a virtual scene and an actual scene according to an embodiment of the present application is shown;
[0079] Figure 8 A block diagram of a computer device according to an embodiment of the present application is shown;
[0080] Fig. 9 A block diagram of a computer device according to another embodiment of the present application is shown. [Specific implementation method]
[0081] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0082] Figure 1 A flow chart of an intelligent driving vehicle testing method according to an embodiment of the present application is shown.
[0083] like Figure 1 As shown, a smart driving vehicle testing method according to an embodiment of the present application includes:
[0084] Step 102, in response to the intelligent driving test instruction, loading a corresponding virtual driving scene for the intelligent driving vehicle in the actual driving scene.
[0085] The virtual driving scene refers to the scene in which the driving conditions of the intelligent driving vehicle need to be tested, and the actual driving scene is the test site where the intelligent driving vehicle used for the test actually moves. In response to the intelligent driving test instruction, the actual movement of the intelligent driving vehicle in the actual driving scene can be displayed as a virtual image in the virtual driving scene for remote intelligent control of the intelligent driving vehicle. The intelligent control mentioned here includes AI control and / or manual remote control.
[0086] Step 104: Acquire first real-time position information of the intelligent driving vehicle in the actual driving scene.
[0087] Step 106: Determine second real-time position information of the intelligent driving vehicle in the virtual driving scene based on the first real-time position information.
[0088] The position and posture information of the intelligent driving vehicle includes at least its three azimuths and three rotation angles, which reflects the actual driving state of the intelligent driving vehicle. Furthermore, when the intelligent driving vehicle in the actual driving scene is projected into the virtual driving scene, a certain error will be generated due to the projection ability. Therefore, the second real-time position and posture information of the intelligent driving vehicle in the virtual driving scene can be obtained after adjustment based on the first real-time position and posture information of the intelligent driving vehicle in the actual driving scene. In this way, the error caused by data conversion in the process of combining virtual and real can be reduced, which helps to reduce the impact of errors on the test results.
[0089] Step 108: Generate an automatic driving adjustment strategy based on the second real-time posture information, the real-time driving environment information provided by the virtual driving scene, and a preset intelligent driving algorithm.
[0090] Step 110: Controlling the intelligent driving vehicle to execute the automatic driving adjustment strategy in the virtual driving scenario and the actual driving scenario respectively.
[0091] The real-time driving environment information provided by the virtual driving scene includes: scene data measured by on-board sensors such as laser radar, millimeter-wave radar, ultrasonic radar, and camera of the historical vehicle in the historical actual scene to which the virtual driving scene belongs; all relevant information of static and dynamic traffic participants in the historical vehicle and the historical actual scene, such as the ID, category, outer dimensions, scene entry time, scene exit time, motion trajectory, heading angle, etc. of the traffic participant; driving data changes of the drive system, brake system, and steering system of the historical vehicle in the historical actual scene, such as changes in brake valve status, steering wheel angle, throttle opening, etc. Among them, the historical vehicle and the intelligent driving vehicle are the same model vehicle.
[0092] In general, the real-time driving environment information provided by the virtual driving scene covers the self-situation and possible environmental information that the intelligent driving vehicle used in the test should have in this test. Therefore, combined with the accurate real-time position information of the intelligent driving vehicle in the virtual driving scene, the automatic driving adjustment strategy that the intelligent driving vehicle should execute in the current virtual driving scene can be calculated through the preset intelligent driving algorithm.
[0093] Furthermore, executing the autonomous driving adjustment strategy includes: sending control instructions required for the autonomous driving adjustment strategy to the driving system, braking system and steering system of the intelligent driving vehicle; and sending warning instructions to the warning module of the intelligent driving vehicle.
[0094] The above technical solution, based on combining virtual driving scenes with actual driving scenes for remote testing, does not need to replicate the actual driving scenes that are consistent with the virtual driving scenes. Instead, it only needs to control the intelligent driving vehicle in an ordinary test site to move according to the real-time driving environment information provided by the virtual driving scene, and the same result as controlling the intelligent driving vehicle to drive in the actual scene corresponding to the virtual driving scene can be obtained. In other words, there is no need to replicate complex scenes for intelligent driving vehicles, but to allow them to obtain the data of complex scenes in ordinary actual driving scenes, so as to achieve the effect of testing intelligent driving vehicles in complex scenes.
[0095] At the same time, this application also takes into account the differences between virtual driving scenarios and actual driving scenarios. After correcting the first real-time posture information of the intelligent driving vehicle in the actual driving scenario, the second real-time posture information of the intelligent driving vehicle in the virtual driving scenario is obtained, so as to reduce the error caused by data conversion between the virtual driving scene and the actual driving scene during the test process and improve the accuracy of the test results.
[0096] In one possible design, a virtual-reality combined simulation system may be built for the intelligent driving vehicle.
[0097] Alternatively, if Figure 2 As shown, the virtual-reality combined simulation system includes a host computer, a rapid prototype machine, the intelligent driving vehicle, a combined navigation module and a communication module, wherein the host computer, the rapid prototype machine, the intelligent driving vehicle and the combined navigation module are connected via the communication module.
[0098] The host computer is used to load the virtual driving scene file reproduced by the real vehicle environment data and display the virtual driving scene corresponding to the virtual driving scene file, and upload the information involved in the virtual driving scene file to the rapid prototyping machine. The virtual driving scene file includes environmental information and motion-related information of historical traffic participants in the virtual driving scene, and the motion-related information includes but is not limited to the motion status and position of historical traffic participants.
[0099] The combined navigation module is used to locate the intelligent driving vehicle, that is, to obtain real-time positioning information and send the positioning information to the rapid prototyping machine.
[0100] The rapid prototyping machine is used to run the preset intelligent driving algorithm based on the information provided by the host computer, the intelligent driving vehicle and the combined navigation module, and control each target execution mechanism of the intelligent driving vehicle to execute the corresponding automatic driving adjustment strategy based on the operation result.
[0101] In one possible design, step 106 includes: determining second real-time posture information of the intelligent driving vehicle in the virtual driving scene based on the posture offset information between the actual driving scene and the virtual driving scene and the first real-time posture information.
[0102] It should be noted that the posture information includes three azimuth angles and three rotation angles of the vehicle. The posture offset information mentioned here refers to the posture offset information corresponding to each numerical value of the three azimuth angles and the three rotation angles, wherein the posture offset information corresponding to each numerical value is different, and respectively represents the degree of influence of the corresponding numerical value on the conversion between the actual driving scene and the virtual driving scene.
[0103] In a possible design, when loading a corresponding virtual driving scene for an intelligent driving vehicle in an actual driving scene, it is first necessary to determine the posture offset information between the actual driving scene and the virtual driving scene.
[0104] Optionally, the method of determining the posture offset information between the actual driving scene and the virtual driving scene includes:
[0105] First, first initial posture information of the intelligent driving vehicle in the actual driving scene and second initial posture information in the virtual driving scene are obtained; then, the posture offset information of the intelligent driving vehicle between the actual driving scene and the virtual driving scene is determined by the difference between the first initial posture information and the second initial posture information.
[0106] That is to say, the difference between the posture information of the intelligent driving vehicle in the actual driving scene and the posture information in the virtual driving scene at the beginning of the test is directly used as the posture offset information between the actual driving scene and the virtual driving scene.
[0107] Optionally, the method of determining the posture offset information between the actual driving scene and the virtual driving scene further includes:
[0108] First, obtain the first initial posture information of the intelligent driving vehicle in the actual driving scene and the second initial posture information in the virtual driving scene; and obtain the difference between the first initial posture information and the second initial posture information; then, based on the type of the virtual driving scene, determine the first difference adjustment coefficient corresponding to the virtual driving scene; based on the difference between the first initial posture information and the second initial posture information, and the first difference adjustment coefficient, determine the posture offset information of the intelligent driving vehicle between the actual driving scene and the virtual driving scene.
[0109] Specifically, the difference between the position information of the intelligent driving vehicle in the actual driving scene and the position information of the virtual driving scene at the beginning of the test is first obtained. Furthermore, since there are many kinds of virtual driving scenes, the more complex the virtual driving scene is, the greater the difference between it and the actual driving scene presented at the test site. Therefore, the virtual driving scenes can be classified according to their complexity, so that the same type of virtual driving scenes have similar differences from the actual driving scenes presented at the test site.
[0110] Furthermore, the degree of difference can be used as the first difference adjustment coefficient corresponding to the virtual driving scene. That is, the larger the first difference adjustment coefficient corresponding to the type of the virtual driving scene is, the greater the difference between the virtual driving scene and the actual driving scene is, the greater the error that may be generated by the data conversion between the two, and the greater the space that needs to be adjusted for the difference between the first initial posture information and the second initial posture information.
[0111] Optionally, the first difference adjustment coefficient is greater than 0 and less than 1.
[0112] Optionally, the posture offset information between the actual driving scene and the virtual driving scene is the product of a difference between the first initial posture information and the second initial posture information and the first difference adjustment coefficient.
[0113] like Figure 3 As shown, the method of determining the posture offset information between the actual driving scene and the virtual driving scene also includes:
[0114] Step 302: Acquire first initial position information of the intelligent driving vehicle in the actual driving scene and second initial position information of the intelligent driving vehicle in the virtual driving scene.
[0115] Step 304: Obtain a difference between the first initial posture information and the second initial posture information.
[0116] Step 306: Determine the scene difference between the actual driving scene and the virtual driving scene based on the preset scene coefficients corresponding to the actual driving scene and the virtual driving scene respectively.
[0117] Specifically, the preset scenario system of each scenario represents the compatibility of the scenario with other scenarios, and the difference between the preset scenario coefficients corresponding to the actual driving scenario and the virtual driving scenario reflects the difference between the actual driving scenario and the virtual driving scenario.
[0118] Optionally, the scene difference between the actual driving scene and the virtual driving scene is determined by using the difference between the preset scene coefficients corresponding to the actual driving scene and the virtual driving scene respectively.
[0119] Step 308: Determine a second difference adjustment coefficient based on the scene difference.
[0120] The second difference adjustment coefficient is used to reflect the size of the space that needs to be adjusted for the difference between the first initial posture information and the second initial posture information. In fact, the higher the scene difference between the actual driving scene and the virtual driving scene, the more complex the virtual driving scene is, and the greater the difference between it and the actual driving scene presented at the test site, the greater the error corresponding to the conversion of the initial posture information of the two.
[0121] Optionally, a square root of the scene difference is determined as a second difference adjustment coefficient to reduce an entropy value in a process of calculating the scene difference.
[0122] Step 310: Determine the posture offset information of the intelligent driving vehicle between the actual driving scene and the virtual driving scene based on the difference between the first initial posture information and the second initial posture information, and the second difference adjustment coefficient.
[0123] Optionally, the second difference adjustment coefficient is greater than 0 and less than 1.
[0124] Optionally, the posture offset information between the actual driving scene and the virtual driving scene is the product of a difference between the first initial posture information and the second initial posture information and the second difference adjustment coefficient.
[0125] In summary, the technical problem to be solved by the present invention is to provide a method and system for in-the-loop testing of a whole vehicle of an intelligent driving vehicle based on the injection of real vehicle environment data, aiming to carry out in-the-loop testing of a whole vehicle of an intelligent driving vehicle based on the injection of real vehicle data in a low-cost, high-efficiency, high-reliability and wide-application manner, and meet the technical requirements of high authenticity, good reliability and strong repeatability of in-the-loop testing of intelligent driving vehicles.
[0126] To achieve the above object, the present invention provides an intelligent driving vehicle vehicle-in-the-loop test system based on real vehicle environment data injection, including a host computer, a rapid prototyping machine, a combined navigation module and a test vehicle (i.e., the aforementioned intelligent driving vehicle). The architecture diagram of the system is as shown in the aforementioned Figure 2 As shown, the combined navigation module is used to update the current position of the test vehicle, the host computer is used to load the virtual historical scene file and visualize the test scene content in real time, the rapid prototype is used to receive the test vehicle positioning, motion information and the traffic participant position, motion status, map and other environmental information in the scene file, and the test vehicle executes the control instructions from the rapid prototype to achieve driving, parking, parking and other actions.
[0127] The specific steps of the system operation are as follows:
[0128] S1: Vehicle-in-the-loop test preparation;
[0129] S2: In-loop testing and obtaining control instructions based on the current virtual scene;
[0130] S3: In-loop testing and updating of the virtual vehicle pose in the virtual scene;
[0131] S4: Repeat test;
[0132] S5: New scene testing.
[0133] Optionally, step S1 includes the following steps:
[0134] S11: Configure the host computer, rapid prototype machine, vehicle, integrated navigation and communication modules, and synchronize test data in real time;
[0135] S12: Select and load a historical scene file reproduced based on real vehicle environment data;
[0136] S13: Configure a CAN communication module in the rapid prototype machine to collect information including but not limited to chassis, body, and positioning information;
[0137] S14: Load and configure the positioning calibration module in the rapid prototype machine;
[0138] S15: Load the intelligent driving algorithm module into the rapid prototype machine, compile and generate a real-time file, and set the rapid prototype machine to the startup mode.
[0139] Furthermore, the real-time synchronization test data in step S11 depends on communication methods such as CAN network, cellular network and Ethernet. The network topology diagram is as follows: Figure 4 As shown, the specific steps include:
[0140] S111: Connect the host computer and the rapid prototype machine by Ethernet communication, and connect the rapid prototype machine with the chassis and the body through the CAN network;
[0141] S112: using the communication module to receive the real-time position of the vehicle from the vehicle remote monitoring center and send it to the rapid prototype machine;
[0142] S113: Solve the data congestion problem caused by multi-channel data parallel transmission and meet the real-time synchronous transmission requirements.
[0143] Specifically, the method for solving the data congestion problem in step S113 includes but is not limited to external devices such as a network bridge.
[0144] Furthermore, the historical scene file used in step S12 is a scene file based on on-board sensor collection, host computer offline processing and online playback. The data in this file comes from a real driving environment. The on-board sensors include but are not limited to lidar, millimeter wave radar, ultrasonic radar, camera, etc. The file also records all scene information of the vehicle, static and dynamic traffic participants, including but not limited to the ID, category, external dimensions, scene entry time, scene exit time, motion trajectory, heading angle, etc. of the traffic participants.
[0145] Furthermore, the positioning information collected in step S13 can be run in the scene model in the host computer to achieve the purpose of receiving and updating the positioning information of the vehicle in the virtual scene in real time, so that the system has a good scene visualization effect; on the other hand, the positioning information can be used in the regulation and control module to design and implement intelligent driving functions. In addition, the vehicle body information includes but is not limited to the wiper status, seat belts and other vehicle body conditions that can be used in the decision-making module.
[0146] Furthermore, step S14 can realize that the test experiment can be carried out in any venue. The test scenarios include but are not limited to the scenarios required by standards and regulations and the scenarios reproduced by real vehicle data. The specific steps are as follows:
[0147] S141: Read the current vehicle position and input it, and convert the longitude and latitude coordinates into the coordinates (X, Y, Yaw, V) of the high-precision map through coordinate transformation. X ,V Y ,Psi);
[0148] S142: Obtain the historical position of the vehicle in the test scene (X l ,Y l ,Yaw l ,V Xl ,V Yl ,Psi l ) and configure it as the target pose (X i ,Y i ,Yaw i ,V Xi ,V Yi ,Psi i , i=1, 2, 3, ...);
[0149] S143: Calculate the actual position and posture (X, Y, Yaw, V) of the test vehicle at the initial moment X ,V Y ,Psi) and target pose (X i ,Y i ,Yaw i ,V Xi ,V Yi ,Psii ) and configure them as pose transformation parameters for real-time calibration of the current pose of the test vehicle in the test scenario. The relationship between the actual pose and the target pose, historical pose and position offset is as follows.
[0150]
[0151] Optionally, step S2 includes the following steps:
[0152] S21: Start the rapid prototype machine, run the scene software in the host computer, and the host computer sends virtual scene information to the rapid prototype machine;
[0153] S22: The driver drives the test vehicle according to the test requirements, and the rapid prototype calculates and generates warnings and control instructions and sends them to the test vehicle.
[0154] Furthermore, the virtual scene information in step S21 can be sent to the rapid prototyping machine via the API of the scene software via Ethernet.
[0155] Furthermore, the warning instruction generated in step S22 is executed by the cockpit display screen or buzzer as an audible or visual warning, and the control instructions include but are not limited to acceleration, braking, steering, etc.
[0156] Preferably, step S3 comprises the following steps:
[0157] S31: The test vehicle receives control instructions from the rapid prototype machine and forwards them to each target mechanism;
[0158] S32: Each target actuator executes the control instruction and feeds back the state signal of the target actuator to the control module in the rapid prototyping machine;
[0159] S33: The rapid control prototype receives the new ego-vehicle posture at the current moment, and forwards it to the virtual scene interface and planning module in the host computer at the same time, so as to achieve the purpose of real-time updating of the current posture of the virtual ego-vehicle model and real-time planning;
[0160] S34: The system repeats steps S31 to S33 until the test vehicle stops or fails to meet the test requirements and terminates this round of testing;
[0161] Furthermore, the target mechanism in step S31 includes but is not limited to a driving system, a braking system and a steering system.
[0162] Furthermore, the target mechanism feedback signal in step S32 includes but is not limited to brake valve status, steering wheel angle, throttle opening, etc.
[0163] Furthermore, the new vehicle position in step S33 is the coordinates (X i ,Y i ,Yaw i ,V Xi ,V Yi ,Psi i ).
[0164] Optionally, step S4 is used to repeat the test steps under the same test case, that is, the historical posture in step S142 has not changed, and specifically includes the following steps:
[0165] S41: placing the real vehicle back into the real scene corresponding to the virtual scene to be tested, and resetting the rapid prototype;
[0166] S42: Repeat steps S2 to S3, re-test and record the test results.
[0167] Furthermore, the corresponding real scene mentioned in step S41 refers to a scene similar to the previous test experiment. The positioning of repeated experiments is not required to be strictly consistent. The positioning error can be overcome by the positioning calibration module of step S41, and resetting the prototype will not affect the pre-configured parameters in the positioning calibration module.
[0168] Preferably, step S5 is used to repeat the experiment under different test cases, that is, the historical pose (X l ,Y l ,Yaw l ,V Xl ,V Yl ,Psi l ) has changed, including the following steps:
[0169] S51: placing the actual vehicle in a new test scene and re-execute step S14;
[0170] S52: Repeat steps S2-S3, re-test and record the test results.
[0171] Furthermore, the purpose of updating the scene to be tested in step S51 is to update the historical pose.
[0172] Therefore, virtual test scenarios can save scene construction costs and ensure test safety; the tests involve actual driving scenarios with high complexity, strong continuity, and high realism, which can improve the authenticity of scene data; the scenes are highly reproducible and repeatable, and test scenarios can be obtained at low cost; the actual test scenes do not need to be completely consistent with the scene files, only the positioning and calibration modules need to be configured, which saves scene reproduction costs to a certain extent; the architecture of host computer-rapid prototype machine-real vehicle can improve the reliability of test results while ensuring the real-time nature of the tests.
[0173] Specifically, the present invention can use virtual test scenarios instead of actual test scenarios, which can save scenario design costs on the one hand, and greatly improve the safety of scenario testing on the other hand. This method only requires configuring scenario elements in the scenario software to complete various test scenarios, making the scenario building work simple and easy to use.
[0174] The virtual scene used in the present invention is derived from the driving environment data collected by the real vehicle, and has the characteristics of being continuous, complex, and in line with real driving habits. This method can avoid the problem of single and overly idealized scenes in standard design scenes as much as possible, and is conducive to simulation testing to obtain the conditions required for vehicle driving in non-continuous scenes.
[0175] The scene data used in the present invention is derived from the on-board sensor, which can realize low-cost collection of scene data, and can obtain a large amount of scene data without the coordination of the vehicle, road and cloud, which is suitable for the current stage of development of autonomous driving technology. In addition, the present invention is provided with a positioning calibration module, which makes it unnecessary for the vehicle to be limited to a test site that is exactly the same as the scene file during testing. It only requires that the site is empty and meets the conditions required for testing, and the posture transformation parameters are configured when used. At the same time, the present invention adopts the architecture scheme of host computer-rapid prototype machine-real vehicle, which can overcome the data delay problem in the communication transmission chain to a certain extent, and provide good real-time communication conditions for the system. In addition, the real vehicle application method better avoids the shortcomings of insufficient reliability of bench experiments.
[0176] In one embodiment of the present application, the reliability of ADAS functions can be tested using a vehicle-in-the-loop test method and system for intelligent driving vehicles. The experimental vehicle used is a bus. The transmission equipment on the bus includes a forward binocular camera, left and right blind spot detection cameras, left and right blind spot image acquisition cameras, ultrasonic radar, integrated navigation, bridge, gateway, 5G communication module, baseline real-time machine and test PC. The network topology diagram of the above equipment is as follows: Figure 5 shown.
[0177] In the preparation phase of vehicle-in-the-loop testing, the test PC, baseline, a certain type of bus instrumentation and other components, combined navigation and 5G communication modules are first configured to meet the needs of real-time synchronization of test data. Among them, the test PC and baseline use Ethernet communication to achieve data connection, and the baseline uses CAN communication with the gateway on the bus, the forward binocular camera, and the left and right blind spot detection cameras. The gateway on the bus is used to forward the status of body equipment such as seat belts and wipers, as well as ultrasonic radar warnings and working status, 5G communication content, etc. The 5G communication module receives the real-time vehicle posture from the vehicle remote monitoring center and sends it to the baseline. In order to avoid data congestion caused by multi-channel data parallel transmission, a bridge device is installed in front of the baseline parallel CAN interface to process data in real time.
[0178] After the equipment is configured, you need to select and load the historical scene file based on the actual vehicle environment data. Figure 6 As shown, it includes a test vehicle, traffic participants and a high-precision map. The traffic participants are composed of two pedestrians O1 and O2 and a passenger car. There is a road and two lanes in the high-precision map. This scene file belongs to the traffic participant result after the fusion of the on-board laser radar and camera. It can output the real-time posture of the traffic participants and scene information such as lane width, number of lanes, road, and traffic light position.
[0179] After determining the scene file, it is necessary to combine the virtual scene and the actual test scene to complete the real-time vehicle posture positioning and scene update. Figure 7 As shown in Figure 2, the initial position of the test vehicle in the virtual historical scene is (X l ,Y l ,Yaw l ,V Xl ,V Yl ,Psi l ), the initial pose in the actual scene is (X, Y, Yaw, V X ,V Y ,Psi), ,the position offset (a1, a2, a3, a4, a5, a6) can be calculated as follows.
[0180]
[0181] Therefore, the instantaneous ego-vehicle posture after calibration by the positioning calibration module can be calculated as follows.
[0182]
[0183] Where i is the true value at the current moment, and i-1 represents the previous moment. After configuring the positioning calibration module, the intelligent driving algorithm to be tested needs to be loaded into the baseline.
[0184] In the loop test and obtaining control instructions based on the current virtual scene, the driver first drives the test vehicle to the test site, then starts the baseline and runs the scenario software on the test PC. Finally, the driver drives the test vehicle according to the test requirements, and the baseline calculates and generates warnings and control instructions in real time and sends them to the vehicle chassis. Among them, the virtual scene information is generated by the scene software API transmitted to the baseline via Ethernet, the warning instructions are executed by the cockpit display or buzzer to perform sound and light warnings, and the control instructions include acceleration, braking, steering and other instructions.
[0185] In the loop test and updating of the virtual vehicle posture in the virtual scene, the test vehicle first receives the control command from the baseline and forwards it to each target mechanism. Then the baseline receives the new self-vehicle posture at the current moment and forwards it to the virtual scene interface and planning module in the test PC at the same time, which is used to update the current posture and real-time planning of the virtual self-vehicle model in real time. During this process, the system continuously performs this operation until the test vehicle stops or fails to meet the test requirements and terminates this round of testing. Among them, the vehicle-side systems involved in the execution include the drive system, brake system and steering system, and the target mechanism feedback signals include the brake valve status, steering wheel angle and throttle opening.
[0186] During the repeated testing phase, the position offset in the positioning calibration module does not change. It is only necessary to place the actual vehicle back in the real scene corresponding to the virtual scene to be tested, reset the rapid prototype, repeat the in-loop test and obtain the control command phase based on the current virtual scene, and repeat the in-loop test and update the virtual vehicle posture phase in the virtual scene.
[0187] In the new scene testing phase, the position offset in the positioning and calibration module has changed. It is necessary to first place the real vehicle in the new test scene or update the historical posture in the test scene, then reconfigure the positioning and calibration module, and finally repeat the in-loop test and obtain the control command based on the current virtual scene and the in-loop test and update the virtual vehicle posture in the virtual scene.
[0188] It should be noted that the data acquisition device used in the present invention includes but is not limited to laser radar, millimeter wave radar, ultrasonic radar, monocular camera, binocular camera, etc.; the high-precision map format used in the present invention includes but is not limited to OpenDrive, OpenStreetMap, HERE HD Live Map, etc.; the scene file format used in the present invention includes but is not limited to OpenScenario; the applicable scenarios of the present invention include but are not limited to open roads, closed parks, test roads, etc.; the self-driving cars mentioned in the present invention include but are not limited to passenger cars, commercial vehicles, engineering vehicles, unmanned vehicles, etc.; the scene design or simulation software used in the present invention includes but is not limited to CARLA, PreScan, RoadRunner, Matlab / Simulink, etc.; the rapid control prototyping machine used in the present invention includes but is not limited to Baseline, Performance, dSpace, etc.
[0189] The present application provides an intelligent driving vehicle testing device, comprising:
[0190] A driving scenario configuration unit, configured to load a corresponding virtual driving scenario for an intelligent driving vehicle in an actual driving scenario in response to an intelligent driving test instruction;
[0191] A first real-time position and posture information acquisition unit, used to acquire first real-time position and posture information of the intelligent driving vehicle in the actual driving scene;
[0192] A second real-time posture information determining unit, configured to determine second real-time posture information of the intelligent driving vehicle in the virtual driving scene based on the first real-time posture information;
[0193] an automatic driving adjustment strategy generating unit, configured to generate an automatic driving adjustment strategy based on the second real-time posture information, the real-time driving environment information provided by the virtual driving scene, and a preset intelligent driving algorithm;
[0194] The automatic driving adjustment strategy execution unit is used to control the intelligent driving vehicle to execute the automatic driving adjustment strategy in the virtual driving scene and the actual driving scene respectively.
[0195] In one embodiment of the present application, optionally, the second real-time posture information determination unit is used to:
[0196] Based on the posture offset information between the actual driving scene and the virtual driving scene and the first real-time posture information, second real-time posture information of the intelligent driving vehicle in the virtual driving scene is determined.
[0197] In one embodiment of the present application, optionally, the device further includes:
[0198] An initial posture information acquisition unit, configured to acquire, when loading a corresponding virtual driving scene for an intelligent driving vehicle in an actual driving scene, first initial posture information of the intelligent driving vehicle in the actual driving scene and second initial posture information in the virtual driving scene;
[0199] The first determining unit is used to determine the posture offset information of the intelligent driving vehicle between the actual driving scene and the virtual driving scene based on the difference between the first initial posture information and the second initial posture information.
[0200] In one embodiment of the present application, optionally, the device further includes:
[0201] An initial posture information acquisition unit, configured to acquire, when loading a corresponding virtual driving scene for an intelligent driving vehicle in an actual driving scene, first initial posture information of the intelligent driving vehicle in the actual driving scene and second initial posture information in the virtual driving scene;
[0202] A difference acquisition unit, used to acquire a difference between the first initial posture information and the second initial posture information;
[0203] A first difference adjustment coefficient determining unit, configured to determine a first difference adjustment coefficient corresponding to the virtual driving scene based on the type of the virtual driving scene;
[0204] The second determination unit is used to determine the posture offset information of the intelligent driving vehicle between the actual driving scene and the virtual driving scene based on the difference between the first initial posture information and the second initial posture information, and the first difference adjustment coefficient.
[0205] In one embodiment of the present application, optionally, the device further includes:
[0206] An initial posture information acquisition unit, configured to acquire, when loading a corresponding virtual driving scene for an intelligent driving vehicle in an actual driving scene, first initial posture information of the intelligent driving vehicle in the actual driving scene and second initial posture information in the virtual driving scene;
[0207] A difference acquisition unit, used to acquire a difference between the first initial posture information and the second initial posture information;
[0208] a scene difference determination unit, configured to determine a scene difference between the actual driving scene and the virtual driving scene based on preset scene coefficients corresponding to the actual driving scene and the virtual driving scene respectively;
[0209] a second difference adjustment coefficient determining unit, configured to determine a second difference adjustment coefficient based on the scene difference;
[0210] The third determination unit is used to determine the posture offset information of the intelligent driving vehicle between the actual driving scene and the virtual driving scene based on the difference between the first initial posture information and the second initial posture information, and the second difference adjustment coefficient.
[0211] In one embodiment of the present application, optionally, the driving scene configuration unit includes:
[0212] A virtual-reality combined simulation system construction unit is used to build a virtual-reality combined simulation system for the intelligent driving vehicle, wherein the virtual-reality combined simulation system includes a host computer, a rapid prototyping machine, the intelligent driving vehicle, a combined navigation module and a communication module.
[0213] The host computer is used to load a virtual driving scene file reproduced by real vehicle environment data and display the virtual driving scene corresponding to the virtual driving scene file, wherein the virtual driving scene file includes environment information and movement association information of historical traffic participants in the virtual driving scene;
[0214] The combined navigation module is used to locate the intelligent driving vehicle;
[0215] The rapid prototyping machine is used to run the preset intelligent driving algorithm based on the information provided by the host computer, the intelligent driving vehicle and the integrated navigation module, and control each target actuator of the intelligent driving vehicle to execute the corresponding automatic driving adjustment strategy based on the running result;
[0216] The host computer, the rapid prototype machine, the intelligent driving vehicle and the combined navigation module are connected via the communication module.
[0217] In one embodiment of the present application, optionally, the autonomous driving adjustment strategy execution unit is used to:
[0218] The control instructions required for the automatic driving adjustment strategy are sent to the driving system, braking system and steering system of the intelligent driving vehicle; and the warning instructions are sent to the warning module of the intelligent driving vehicle.
[0219] The device uses any one of the solutions described in the above embodiments, and therefore has all the above technical effects, which will not be repeated here.
[0220] In addition, in one embodiment, the present application provides a computer device, which may be a server, and its internal structure diagram may be as shown in FIG. Figure 8As shown. The computer device includes a processor, a memory, a network interface and a database connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile and / or volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface of the computer device is used to communicate with an external client via a network connection. When the computer program is executed by the processor, the method described in any of the above embodiments can be implemented.
[0221] In one embodiment, the present application further provides a computer device, which may be a client, and its internal structure diagram may be as follows: Fig. 9 As shown. The computer device includes a processor, a memory, a network interface, a display screen and an input device connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface of the computer device is used to communicate with an external server via a network connection. When the computer program is executed by the processor, the method described in any of the above embodiments can be implemented.
[0222] Any of the above-mentioned computer devices in the embodiments of the present application may exist in various forms, including but not limited to:
[0223] (1) Mobile communication devices: These devices are characterized by having mobile communication functions and their main purpose is to provide voice and data communications. These terminals include: smart phones (such as iPhone), multimedia phones, functional phones, and low-end phones.
[0224] (2) Ultra-mobile personal computer devices: These devices fall into the category of personal computers, have computing and processing capabilities, and generally also have mobile Internet access features. These terminals include: PDA, MID and UMPC devices, such as iPad.
[0225] (3) Portable entertainment devices: These devices can display and play multimedia content. They include audio and video players (such as iPods), handheld game consoles, e-books, as well as smart toys, wearable devices, and portable car navigation devices.
[0226] (4) Server: A device that provides computing services. The server consists of a processor, hard disk, memory, system bus, etc. The server is similar to a general computer architecture, but because it needs to provide highly reliable services, it has higher requirements in terms of processing power, stability, reliability, security, scalability, and manageability.
[0227] (5) Other electronic devices with data interaction functions.
[0228] In addition, an embodiment of the present application provides a computer-readable storage medium storing computer-executable instructions, wherein the computer-executable instructions are used to perform the following steps:
[0229] In response to the intelligent driving test instruction, loading a corresponding virtual driving scene for the intelligent driving vehicle in the actual driving scene;
[0230] Acquire first real-time position information of the intelligent driving vehicle in the actual driving scene;
[0231] Based on the first real-time position information, determining second real-time position information of the intelligent driving vehicle in the virtual driving scene;
[0232] Generate an automatic driving adjustment strategy based on the second real-time posture information, the real-time driving environment information provided by the virtual driving scene, and a preset intelligent driving algorithm;
[0233] The intelligent driving vehicle is controlled to execute the automatic driving adjustment strategy in the virtual driving scenario and the actual driving scenario respectively.
[0234] It should be noted that the above functions or steps that can be implemented by the computer-readable storage medium or computer device can refer to the relevant description in the aforementioned method embodiment. In order to avoid repetition, they will not be described one by one here.
[0235] The technical solution of the present application is described in detail above in conjunction with the accompanying drawings. Through the technical solution of the present application, on the basis of combining the virtual driving scene with the actual driving scene for remote testing, there is no need to replicate the actual driving scene that is consistent with the virtual driving scene, but only needs to control the intelligent driving vehicle in the ordinary test site to move according to the real-time driving environment information provided by the virtual driving scene, and the same result of controlling the intelligent driving vehicle to drive in the actual scene corresponding to the virtual driving scene can be obtained. In other words, there is no need to replicate complex scenes for intelligent driving vehicles, but to allow them to obtain the data of complex scenes in ordinary actual driving scenes, so as to achieve the effect of testing intelligent driving vehicles in complex scenes. At the same time, the difference between virtual driving scenes and actual driving scenes is also taken into account in the present application. After correcting the first real-time posture information of the intelligent driving vehicle in the actual driving scene, the second real-time posture information of the intelligent driving vehicle in the virtual driving scene is obtained, so as to reduce the error caused by the data conversion between the virtual driving scene and the actual driving scene during the test, and improve the accuracy of the test results.
[0236] The word "if" as used herein may be interpreted as "at the time of" or "when" or "in response to determining" or "in response to detecting", depending on the context. Similarly, the phrases "if it is determined" or "if (stated condition or event) is detected" may be interpreted as "when it is determined" or "in response to determining" or "when detecting (stated condition or event)" or "in response to detecting (stated condition or event)", depending on the context.
[0237] The terms used in the embodiments of the present application are only for the purpose of describing specific embodiments, and are not intended to limit the present application. The singular forms "a", "said" and "the" used in the embodiments of the present application and the appended claims are also intended to include plural forms, unless the context clearly indicates other meanings.
[0238] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.
[0239] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of hardware plus software functional units.
[0240] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).
[0241] The embodiments described above are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included in the protection scope of the present invention.
Claims
1. A method for testing an intelligent driving vehicle, characterized in that: include: In response to the intelligent driving test instruction, loading a corresponding virtual driving scene for the intelligent driving vehicle in the actual driving scene; Acquire first real-time position information of the intelligent driving vehicle in the actual driving scene; Based on the first real-time position information, determining second real-time position information of the intelligent driving vehicle in the virtual driving scene; Generate an automatic driving adjustment strategy based on the second real-time posture information, the real-time driving environment information provided by the virtual driving scene, and a preset intelligent driving algorithm; The intelligent driving vehicle is controlled to execute the automatic driving adjustment strategy in the virtual driving scenario and the actual driving scenario respectively.
2. The method according to claim 1, characterized in that: The determining, based on the first real-time posture information, second real-time posture information of the intelligent driving vehicle in the virtual driving scene includes: Based on the posture offset information between the actual driving scene and the virtual driving scene and the first real-time posture information, second real-time posture information of the intelligent driving vehicle in the virtual driving scene is determined.
3. The method according to claim 2, characterized in that When loading the corresponding virtual driving scene for the intelligent driving vehicle in the actual driving scene, it also includes: Acquire first initial position information of the intelligent driving vehicle in the actual driving scene and second initial position information of the intelligent driving vehicle in the virtual driving scene; The difference between the first initial posture information and the second initial posture information is used to determine the posture offset information of the intelligent driving vehicle between the actual driving scene and the virtual driving scene.
4. The method according to claim 2, characterized in that: When loading the corresponding virtual driving scene for the intelligent driving vehicle in the actual driving scene, it also includes: Acquire first initial position information of the intelligent driving vehicle in the actual driving scene and second initial position information of the intelligent driving vehicle in the virtual driving scene; Obtaining a difference between the first initial posture information and the second initial posture information; Determining a first difference adjustment coefficient corresponding to the virtual driving scene based on the type of the virtual driving scene; Based on the difference between the first initial posture information and the second initial posture information, and the first difference adjustment coefficient, the posture offset information of the intelligent driving vehicle between the actual driving scene and the virtual driving scene is determined.
5. The method according to claim 2, characterized in that: When loading the corresponding virtual driving scene for the intelligent driving vehicle in the actual driving scene, it also includes: Acquire first initial position information of the intelligent driving vehicle in the actual driving scene and second initial position information of the intelligent driving vehicle in the virtual driving scene; Obtaining a difference between the first initial posture information and the second initial posture information; Determining a scene difference between the actual driving scene and the virtual driving scene based on preset scene coefficients corresponding to the actual driving scene and the virtual driving scene respectively; Based on the scene difference, determining a second difference adjustment coefficient; Based on the difference between the first initial posture information and the second initial posture information, and the second difference adjustment coefficient, the posture offset information of the intelligent driving vehicle between the actual driving scene and the virtual driving scene is determined.
6. The method according to any one of claims 1 to 5, characterized in that The method of loading a corresponding virtual driving scene for an intelligent driving vehicle in an actual driving scene includes: A virtual-reality combined simulation system is built for the intelligent driving vehicle, wherein the virtual-reality combined simulation system includes a host computer, a rapid prototyping machine, the intelligent driving vehicle, a combined navigation module and a communication module. The host computer is used to load a virtual driving scene file reproduced by real vehicle environment data and display the virtual driving scene corresponding to the virtual driving scene file, wherein the virtual driving scene file includes environment information and movement association information of historical traffic participants in the virtual driving scene; The combined navigation module is used to locate the intelligent driving vehicle; The rapid prototyping machine is used to run the preset intelligent driving algorithm based on the information provided by the host computer, the intelligent driving vehicle and the integrated navigation module, and control each target actuator of the intelligent driving vehicle to execute the corresponding automatic driving adjustment strategy based on the running result; The host computer, the rapid prototype machine, the intelligent driving vehicle and the combined navigation module are connected via the communication module.
7. The method according to claim 6, characterized in that The controlling each target execution mechanism of the intelligent driving vehicle to execute a corresponding automatic driving adjustment strategy includes: Sending control instructions required for the automatic driving adjustment strategy to the driving system, braking system and steering system of the intelligent driving vehicle; Sending a warning instruction to the warning module of the intelligent driving vehicle.
8. An intelligent driving vehicle testing device, characterized in that: include: A driving scenario configuration unit, configured to load a corresponding virtual driving scenario for an intelligent driving vehicle in an actual driving scenario in response to an intelligent driving test instruction; A first real-time position and posture information acquisition unit, used to acquire first real-time position and posture information of the intelligent driving vehicle in the actual driving scene; A second real-time posture information determining unit, configured to determine second real-time posture information of the intelligent driving vehicle in the virtual driving scene based on the first real-time posture information; an automatic driving adjustment strategy generating unit, configured to generate an automatic driving adjustment strategy based on the second real-time posture information, the real-time driving environment information provided by the virtual driving scene, and a preset intelligent driving algorithm; The automatic driving adjustment strategy execution unit is used to control the intelligent driving vehicle to execute the automatic driving adjustment strategy in the virtual driving scene and the actual driving scene respectively.
9. A computer device, characterized in that: include: at least one processor; and, a memory communicatively coupled to the at least one processor; The memory stores instructions executable by the at least one processor, wherein the instructions are configured to execute the method according to any one of claims 1 to 7.
10. A computer-readable storage medium, characterized in that: Computer executable instructions are stored, and the computer executable instructions are used to execute the method according to any one of claims 1 to 7.
Citation Information
Patent Citations
Automatic driving vehicle simulation test method and device, computer device and storage medium
CN111505965A
Large enclosed space immersive driving system and control method
CN113157098A
Vehicle in-loop fusion test system and method
CN114896817A
Automatic driving real vehicle in-the-loop test method and system for customizing virtual obstacle
CN116736735A
Calibration method and device of motion platform, storage medium and vehicle
CN117288488A
