Automatic driving test system, automatic driving algorithm test method and equipment

Abstract

The invention provides an automatic driving test system, and a test method and equipment of an automatic driving algorithm. The system comprises a data playing module, a data simulation module, an automatic driving algorithm module and a test module. And the data playing module plays the test data of the vehicle, wherein the test data comprises n test data corresponding to the n timestamps. And the data simulation module generates a simulated real environment according to the test data. Wherein the vehicle is controlled to run in the simulated real environment according to the first original vehicle state information in the first test data, and the state information of the vehicle is determined. And the automatic driving algorithm module executes an automatic driving algorithm based on the i-th test data and the current state information and outputs an i-th control instruction. And the data simulation module controls the vehicle to run in a simulated real environment based on the ith control instruction and determines the state information of the vehicle. And updating i by using i + 1, and repeating the process until i is equal to n. And the test module determines a test result of the automatic driving algorithm in combination with the state information of the vehicle.

Description

Technical Field

[0001] This application relates to the field of autonomous driving technology, and particularly to an autonomous driving test system, a test method and device for an autonomous driving algorithm. Background Art

[0002] With the rapid development of autonomous driving technology, the importance of autonomous driving algorithm testing has become increasingly prominent. In the autonomous driving development process, Software-In-the-Loop (SIL) testing is a key link for verifying the functions of software algorithms. Without relying on actual hardware, it can verify the functional correctness of autonomous driving algorithms through a simulation environment.

[0003] Currently, there are mainly two ways for SIL testing: real vehicle data playback testing and simulation testing. Real vehicle data playback testing refers to using pre-collected road sensor data for real vehicle playback to test the functions of autonomous driving algorithms in a real environment. However, this method cannot modify the test scenarios and it is difficult to combine the real-time control function results of the algorithms for feedback, which affects the accuracy of testing. Simulation testing constructs test scenarios in a simulation environment, which can flexibly build scenarios, but it is time-consuming and laborious to build scenarios, with low efficiency, and the authenticity of simulation data is poor, which affects the test effect. Summary of the Invention

[0004] In view of this, this application is committed to providing an autonomous driving test system, a test method and device for an autonomous driving algorithm, so as to improve the test efficiency and the accuracy of test results while expanding the test scenarios.

[0005] In a first aspect, this application provides an autonomous driving test system, which includes: a data playback module, a data simulation module, an autonomous driving algorithm module, and a test module;

[0006] The data playback module is used to play test data of a vehicle, where the test data includes n test data corresponding to n timestamps, and the test data is data generated when the vehicle was driving in a real environment in advance;

[0007] The data simulation module is used to generate a simulated real environment according to the test data; control the vehicle to drive in the simulated real environment based on the first original vehicle state information in the first test data; and determine the state information of the vehicle;

[0008] The autonomous driving algorithm module is used to execute an autonomous driving algorithm based on the i-th test data and the current state information of the vehicle obtained from the data simulation module, and output the i-th control instruction, where the i-th test data represents the test data corresponding to the i-th timestamp, and i = 1, 2,..., n;

[0009] The data simulation module is further configured to control the vehicle to travel in the simulated real environment based on the i-th control instruction, and determine the state information of the vehicle;

[0010] Update i with i + 1, and repeat the above process until i = n;

[0011] The testing module is configured to determine the test result of the autonomous driving algorithm based on the state information of the vehicle.

[0012] In a possible implementation manner, the data playback module is configured to play the corresponding n test data in the order of n timestamps, and the i-th test data includes the i-th environmental data and the i-th original vehicle state information;

[0013] The autonomous driving algorithm module is configured to, when receiving the i-th test data corresponding to the i-th timestamp, obtain the current state information of the vehicle from the data simulation module, where the current state information is the latest state information of the vehicle in the simulated real environment corresponding to the i-th timestamp; execute the autonomous driving algorithm based on the i-th environmental data and the current state information of the vehicle, and output the i-th control instruction.

[0014] In a possible implementation manner, when the i-th test data includes the original relative data of the obstacle relative to the vehicle, the data simulation module is further configured to determine the absolute data of the obstacle based on the i-th original vehicle state information and the original relative data in the i-th test data; determine the real relative data of the obstacle relative to the vehicle in the simulated real environment based on the absolute data and the current state information corresponding to the i-th test data, and feed back the real relative data to the autonomous driving algorithm module;

[0015] The autonomous driving algorithm module is configured to update the original relative data in the i-th test data based on the real relative data, and execute the autonomous driving algorithm based on the updated i-th test data and the current state information corresponding to the i-th test data, and output the i-th control instruction.

[0016] In a possible implementation manner, the testing module is configured to determine that the test result of the autonomous driving algorithm is qualified when the state information meets the preset control strategy; otherwise, determine that the test result of the autonomous driving algorithm is unqualified.

[0017] In a possible implementation, the system further includes: a middleware module, configured to receive the test data; convert the data format of the test data to obtain first-converted test data, and send the first-converted test data to the autonomous driving algorithm module; convert the data format of the test data to obtain second-converted test data, and send the second-converted test data to the data simulation module.

[0018] In a possible implementation, the system further includes: a data processing module, configured to obtain the original test data when the vehicle travels in advance; classify and slice the original test data based on test requirements to obtain a plurality of the test data and labels corresponding to the test data; and send the plurality of test data to the data playback module in timestamp order.

[0019] In a possible implementation, the system further includes: a visualization module, configured to render and display a three-dimensional scene of the vehicle traveling in the simulated real environment; or display function buttons for controlling the playback of the test data; or support user interaction operations on the visualization module.

[0020] In a possible implementation, the data playback module is configured to adjust the playback rate of the test data; or pause / resume the playback of the test data; or skip to play the test data.

[0021] In a second aspect, the present application provides a method for testing an autonomous driving algorithm. The method is applied to an autonomous driving algorithm module, and the method includes:

[0022] Obtain the i-th test data of the vehicle, where the i-th test data is data generated when the vehicle travels in a real environment in advance. The test data includes n test data corresponding to n timestamps, and the i-th test data represents the test data corresponding to the i-th timestamp, where i = 1, 2,..., n;

[0023] Obtain the current state information of the vehicle determined by the data simulation module corresponding to the i-th test data;

[0024] Execute an autonomous driving algorithm based on the i-th test data and the current state information, and output an i-th control instruction. The data simulation module is configured to generate a simulated real environment according to the i-th test data; control the vehicle to travel in the simulated real environment based on the first original vehicle state information in the first test data; and determine the state information of the vehicle. In addition, the data simulation module is configured to control the vehicle to travel in the simulated real environment based on the i-th control instruction and determine the state information of the vehicle.

[0025] Update i with i + 1, and repeat the above process until i = n, so that the test module determines the test result of the autonomous driving algorithm based on the state information of the vehicle.

[0026] In a third aspect, the present application provides a test device for an autonomous driving algorithm. The device is applied to an autonomous driving algorithm module, and the device includes:

[0027] A first acquisition unit, configured to acquire the i-th test data of the vehicle, where the i-th test data is data generated by the vehicle during a prior real environment driving. Among them, the test data includes n test data corresponding to n timestamps, and the i-th test data represents the test data corresponding to the i-th timestamp, where i = 1, 2,..., n;

[0028] A second acquisition unit, configured to acquire the current state information of the vehicle determined by the data simulation module corresponding to the i-th test data;

[0029] An algorithm execution unit, configured to execute the autonomous driving algorithm based on the i-th test data and the current state information, and output the i-th control instruction; where the data simulation module is configured to generate a simulated real environment according to the i-th test data; control the vehicle to drive in the simulated real environment based on the first original vehicle state information in the first test data; and determine the state information of the vehicle; and, configured to control the vehicle to drive in the simulated real environment based on the i-th control instruction, and determine the state information of the vehicle;

[0030] An iteration unit, configured to update i with i + 1, and repeat the above process until i = n, so that the test module determines the test result of the autonomous driving algorithm based on the state information of the vehicle.

[0031] In a fourth aspect, the present application provides an electronic device, and the device includes: a memory and a processor;

[0032] The memory is used to store relevant program codes;

[0033] The processor is configured to call the program codes to execute the test method of the autonomous driving algorithm according to any one of the implementation manners in the second aspect above.

[0034] In a fifth aspect, the present application provides a computer-readable storage medium, and the computer-readable storage medium is used to store a computer program, and the computer program is used to execute the test method of the autonomous driving algorithm according to any one of the implementation manners in the second aspect above.

[0035] Sixth aspect, the present application provides a computer program product, the computer program product includes computer programs / instructions, and when the computer programs / instructions are executed by a processor, the test method of the autonomous driving algorithm described in any implementation manner of the second aspect above is implemented.

[0036] In the above implementation manner of the present application, an autonomous driving test system is provided. The system includes: a data playback module, a data simulation module, an autonomous driving algorithm module, and a test module. Among them, during the test, the data playback module can play the test data of the vehicle, and the test data is the data generated when the vehicle travels in the real environment in advance. Among them, the test data includes n test data corresponding to n timestamps. After receiving the test data played by the data playback module, the data simulation module can generate a simulated real environment according to the test data. Among them, when receiving the first test data, that is, the initial test data, the vehicle can be controlled to travel in the simulated real environment according to the first original vehicle state information in the first test data. When the vehicle travels in the simulated real environment, the data simulation module can determine the state information of the vehicle in the simulated real environment, including the position, speed, acceleration, etc. of the vehicle. When receiving the i-th test data, the autonomous driving algorithm module can obtain the current state information of the vehicle from the data simulation module, and based on the i-th test data and the current state information, execute the autonomous driving algorithm and output the i-th control instruction for the vehicle. Among them, the i-th test data represents the test data corresponding to the i-th timestamp, i = 1, 2,..., n. After receiving the i-th control instruction output by the autonomous driving algorithm module, the data simulation module controls the vehicle to travel in the simulated real environment based on the i-th control instruction, and can determine the state information of the vehicle after traveling according to the control instruction. Update i with i + 1, and repeat the above process until i = n, that is, all the test data of the timestamps are played. At this time, the test module can determine the test result of the autonomous driving algorithm in combination with the state information of the vehicle after executing the control instruction. Through the system provided by the present application, the autonomous driving algorithm can be tested in a simulated real environment, the test scenario can be extended, and the test efficiency can be improved. And using the test data of the vehicle traveling in the real environment for testing can ensure the authenticity of the data, thereby improving the accuracy of the test result. The result of the vehicle executing according to the control instruction can also be fed back to the autonomous driving algorithm to complete the closed-loop test, further improving the accuracy of the test result. Description of the Drawings

[0037] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following drawings are only some embodiments provided in the present application. For those of ordinary skill in the art, other drawings can also be obtained according to these drawings.

[0038] Figure 1 Schematic diagram of a structure of an autonomous driving test system provided by an embodiment of the present application.

[0039] Figure 2a Schematic diagram of a navigation map provided by an embodiment of the present application.

[0040] Figure 2b Schematic diagram of a high-precision map provided by an embodiment of the present application.

[0041] Figure 3 Schematic diagram of a structure of another autonomous driving test system provided by an embodiment of the present application.

[0042] Figure 4 Schematic diagram of a structure of yet another autonomous driving test system provided by an embodiment of the present application.

[0043] Figure 5 Schematic diagram of a structure of yet another autonomous driving test system provided by an embodiment of the present application.

[0044] Figure 6 Flowchart of a test method for an autonomous driving algorithm provided by an embodiment of the present application.

[0045] Figure 7 Schematic diagram of a test device for an autonomous driving algorithm provided by an embodiment of the present application.

[0046] Figure 8 Schematic diagram of an electronic device provided by an embodiment of the present application. Detailed implementation manners

[0047] Next, the technical solutions in the embodiments of the present application will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present application. The described embodiments are only exemplary implementation manners of the present application, not all implementation manners. Those skilled in the art can obtain other embodiments without creative efforts in combination with the embodiments of the present application, and these embodiments are also within the protection scope of the present application.

[0048] In the autonomous driving development process, Software-In-the-Loop (SIL) testing is a key link for verifying the functions of software algorithms. Without relying on actual hardware, the functional correctness of autonomous driving algorithms can be verified through a simulation environment.

[0049] At present, there are mainly two ways of SIL testing: real vehicle data playback testing and simulation testing. Real vehicle data playback testing refers to using pre-collected road sensor data for real vehicle playback to test the functions of the autonomous driving algorithm in a real environment. However, this method cannot modify the test scenarios and it is difficult to combine the results of the real-time control function of the algorithm for feedback, which affects the accuracy of the test. Simulation testing constructs test scenarios in a simulation environment, where test scenarios can be flexibly built. However, the construction of scenarios is time-consuming and laborious, with low efficiency, and the authenticity of the simulation data is poor, which affects the test effect.

[0050] Based on this, the embodiments of the present application provide an autonomous driving test system for testing an autonomous driving algorithm, so as to improve the test efficiency and the accuracy of the test results while expanding the test scenarios. Specifically, during the test, the data playback module can play the test data of the vehicle, which is the data generated when the vehicle was driving in a real environment in advance. Among them, the test data includes n test data corresponding to n timestamps. After receiving the test data played by the data playback module, the data simulation module can generate a simulated real environment based on the test data. Among them, when receiving the first test data, that is, the initial test data, the vehicle can be controlled to drive in the simulated real environment according to the first original vehicle state information in the first test data. When the vehicle is driving in the simulated real environment, the data simulation module can determine the state information of the vehicle in the simulated real environment, including the position, speed, acceleration, etc. of the vehicle. When receiving the i-th test data, the autonomous driving algorithm module can obtain the current state information of the vehicle from the data simulation module, and based on the i-th test data and the current state information, execute the autonomous driving algorithm and output the i-th control instruction for the vehicle. Among them, the i-th test data represents the test data corresponding to the i-th timestamp, and i = 1, 2,..., n. After receiving the i-th control instruction output by the autonomous driving algorithm module, the data simulation module controls the vehicle to drive in the simulated real environment based on the i-th control instruction, and can determine the state information of the vehicle after driving according to the control instruction. Update i with i + 1 and repeat the above process until i = n, that is, all the test data of the timestamps are played. At this time, the test module can determine the test result of the autonomous driving algorithm in combination with the state information of the vehicle after executing the control instruction.

[0051] Through the system provided by this application, the autonomous driving algorithm can be tested in a simulated real environment, and the test scenarios can be extended by modifying the parameters of the simulated real environment. Moreover, compared with the environment constructed by pure simulation, creating a simulated real environment can reduce time and improve the test efficiency. Using the test data of the vehicle driving in the real environment for testing can ensure the authenticity of the data, thereby improving the accuracy of the test results. The result of the vehicle's execution according to the control instruction can also be fed back to the autonomous driving algorithm, so that the autonomous driving algorithm module can execute the autonomous driving algorithm in combination with the new state information of the vehicle to complete the closed-loop test and further improve the accuracy of the test results.

[0052] To facilitate the understanding of the technical solution provided by the embodiments of this application, the following will specifically introduce it in combination with the accompanying drawings in the embodiments.

[0053] Specifically, it can be seen Figure 1 as shown in Figure 1 the structural schematic diagram of an autonomous driving test system provided by the embodiments of this application.

[0054] The system 100 includes: a data playback module 101, a data simulation module 102, an autonomous driving algorithm module 103, and a test module 104.

[0055] The data playback module 101 is used to play the test data of the vehicle. Among them, the test data is the data generated when the vehicle travels in the real environment in advance.

[0056] That is to say, the vehicle can be pre-controlled to travel in the real environment, and the vehicle can be controlled to travel according to the behaviors to be tested (such as accelerating, decelerating, turning, etc.), so as to generate the required test data. The test data may include map data when the vehicle travels, perception data (such as obstacles, etc.), and relevant parameters of the vehicle (such as speed, acceleration, etc.). Among them, the test data includes environmental data and original vehicle state information. The environmental data includes, for example, map data when the vehicle travels, perception data (such as obstacles, etc.), and the original vehicle state information includes, for example, relevant parameters of the vehicle (such as speed, acceleration, etc.).

[0057] In a possible implementation manner, the test data may include n test data corresponding to n timestamps respectively. That is, the first timestamp corresponds to the first test data, the second timestamp corresponds to the second test data, the i-th timestamp corresponds to the i-th test data, and so on, where i = 1, 2,..., n, and n represents the total number of timestamps. Among them, the first test data includes the first environmental data and the first original vehicle state information; the second test data includes the second environmental data and the second original vehicle state information, and so on.

[0058] The data playback module 101 can play the test data in the order of timestamps, and each timestamp corresponds to a set of test data. That is to say, the test data is a continuous playback process, and the entire test process ends until the test data playback is completed. Among them, the process of determining the test data corresponding to multiple timestamps can be referred to the subsequent introduction of the data processing module, and will not be elaborated here.

[0059] The data simulation module 102 is used to generate a simulated real environment according to the test data; control the vehicle to drive in the simulated real environment based on the first original vehicle state information in the first test data; and determine the state information of the vehicle in the simulated real environment.

[0060] In a possible implementation, since the test data includes map data, perception data (such as obstacles, etc.), and relevant parameters of the vehicle (such as speed, acceleration, etc.) when the vehicle is driving, when the data simulation module 102 receives the test data played by the data playback module 101, it can generate a simulated real environment according to the test data, so that the vehicle can drive in the simulated real environment, which is used to simulate the real scenario of vehicle driving. Optionally, the data simulation module 102 is used to generate a simulated real environment according to the environmental data in the test data.

[0061] In the initial test process, the first test data corresponding to the first timestamp can be used to initialize the simulated real environment as the initial state of the vehicle driving in the simulated real environment. Therefore, after receiving the first test data, the data simulation module 102 can control the vehicle to drive in the simulated real environment based on the first original vehicle state information in the first test data. When the vehicle is driving in the simulated real environment, the data simulation module 102 can determine the real-time state information of the vehicle, including the position, speed, acceleration, etc. of the vehicle.

[0062] In a possible implementation, the simulated real environment provided by the embodiments of the present application can be a three-dimensional scene that simulates the vehicle driving in the real environment. In the simulated real environment, maps, various behaviors of vehicle models, obstacles, etc. can be displayed. Optionally, the functions of the map engine can be integrated in the data simulation module 102, and these functions include navigation maps and high-precision maps. In the simulated real environment, the navigation map can provide the global navigation path, and the high-precision map can provide lane-level path information. The data simulation module can continuously update the simulated real environment in combination with the test data. For details, please refer to Figure 2a and Figure 2b as shown Figure 2a is a schematic diagram of a navigation map provided by the embodiments of the present application, Figure 2b is a schematic diagram of a high-precision map provided by the embodiments of the present application.

[0063] Since the data simulation module 102 can simulate a real environment based on the test data after receiving the test data, it controls the vehicle to drive in the simulated real environment, which is used to simulate the driving state of the vehicle in the real environment. Based on this, by modifying the parameters of the test data, the created simulated real environment can be modified, and the test scenarios for vehicle driving can be flexibly expanded. Compared with the environment constructed by the pure simulation algorithm, creating a simulated real environment can reduce time, improve the test efficiency, increase the authenticity of the simulation data, and enhance the test effect.

[0064] The automatic driving algorithm module 103 is used to execute the automatic driving algorithm based on the i-th test data and the current state information of the vehicle obtained from the data simulation module, and output the i-th control instruction.

[0065] Since the position and speed of the vehicle may change during the driving process in the simulated real environment, before the automatic driving algorithm module 103 executes the automatic driving algorithm, it can obtain the latest current state information of the vehicle according to the requirements input by the automatic driving algorithm, including the current position, speed or acceleration of the vehicle, etc. Moreover, the data playback module 101 can send the i-th test data corresponding to the i-th time stamp to the automatic driving algorithm module 103 in the order of time stamps. The automatic driving algorithm module 103 can execute the automatic driving algorithm based on the i-th test data and the current state information of the vehicle, and output the i-th control instruction of the vehicle to the data simulation module 102. For example, the i-th control instruction can be to control the vehicle to go straight, turn, accelerate or decelerate, etc.

[0066] After receiving the i-th control instruction output by the automatic driving algorithm module 103, the data simulation module 102 can control the vehicle to drive in the simulated real environment based on the i-th control instruction. After changing the driving state of the vehicle based on the i-th control instruction, the data simulation module 102 can determine the state information of the vehicle. The data simulation module 102 can determine the state information of the vehicle based on the input data according to the kinematic principle.

[0067] After determining the status information of the vehicle after executing the control instruction, the data simulation module 102 can send the status information to the test module 104, and the test module 104 determines the test result of the autonomous driving algorithm based on the status information. That is, the test module 104 determines whether the control of the autonomous driving algorithm meets the requirements based on the status information to determine the test result of the autonomous driving algorithm. Specifically, when the status information of the vehicle conforms to the preset control strategy, the test result of the autonomous driving algorithm is determined to be qualified. When the status information of the vehicle does not conform to the preset control strategy, the test result of the autonomous driving algorithm is determined to be unqualified. Among them, the preset control strategy can be understood as the vehicle avoiding obstacles and driving safely to the preset destination. For example, it can be to decelerate and move away from the obstacle in front, or to steer to avoid the nearby obstacle, etc. That is, based on the control result of the control instruction output by the autonomous driving algorithm, the vehicle's status information is changed to enable the vehicle to avoid obstacles and drive safely.

[0068] In a possible implementation manner, after obtaining the status information of the vehicle, the autonomous driving algorithm module can also determine the test result of the autonomous driving algorithm based on the status information. At this time, it can be understood that the test module and the autonomous driving algorithm module are the same module. It should be noted that the modules provided in the embodiments of the present application are only named based on the implemented functions, and do not limit that different named modules are different modules, that is, different named modules can also be the same module.

[0069] Since there is test data corresponding to multiple timestamps, the data playback module 101 can play the corresponding test data in timestamp order. That is, update i with i + 1 and repeat the above process. That is, the data playback module plays the test data corresponding to the next timestamp (the (i + 1)-th timestamp) to start the next round of testing process based on the test data. After receiving the (i + 1)-th test data corresponding to the (i + 1)-th timestamp, the data simulation module 102 can continue to generate a simulated real environment based on the (i + 1)-th test data, control the vehicle to drive in the simulated real environment, and determine the status information of the vehicle. After receiving the (i + 1)-th test data, the autonomous driving algorithm module 103 can obtain the latest current status information of the vehicle corresponding to the (i + 1)-th test data from the data simulation module 102, execute the autonomous driving algorithm based on the (i + 1)-th test data and the current status information of the corresponding vehicle, and output the (i + 1)-th control instruction to the data simulation module 102. After receiving the (i + 1)-th control instruction, the data simulation module 102 controls the vehicle to drive in the simulated real environment based on the (i + 1)-th control instruction and determines the status information of the vehicle. After determining the status information of the vehicle after executing the control instruction, the data simulation module 102 sends the status information to the test module 104 for the test module 104 to determine the test result of the autonomous driving algorithm.

[0070] Since it may be necessary to test multiple control functions of the autonomous driving algorithm during the entire test process to determine whether the control functions of the autonomous driving algorithm meet the preset control strategy. That is, the autonomous driving algorithm module 103 can output multiple control instructions, and thus multiple state information of the vehicle after the data simulation module 102 controls the vehicle to drive according to the control instructions can be obtained. The test module needs to test each state information. Optionally, when the test module 103 determines that any one of the state information does not meet the preset control strategy, it is determined that the test result of the autonomous driving algorithm is unqualified. In specific implementation, when the data playback module plays the corresponding n test data in the order of n timestamps, where the i-th test data includes the i-th environmental data and the i-th original vehicle state information, when the autonomous driving algorithm module 103 receives the i-th test data corresponding to the i-th timestamp, it can obtain the current state information of the vehicle from the data simulation module. Among them, the current state information is the latest state information of the vehicle corresponding to the i-th timestamp. Since the state information of the vehicle is continuously changing, the data simulation module 102 can generate the latest state information of the vehicle regularly, that is, generate the latest state information of the vehicle at regular intervals. The autonomous driving algorithm module 103 can obtain the latest current state information of the vehicle generated by the data simulation module 102 after receiving the i-th test data. Then, based on the i-th environmental data and the current state information of the vehicle, the autonomous driving algorithm is executed, and the i-th control instruction of the vehicle is output to the data simulation module 102.

[0071] Through the system provided by the embodiments of the present application, the autonomous driving algorithm can be tested in a simulated real environment, and the test scenario can be extended by modifying the parameters of the simulated real environment. Compared with the environment constructed by pure simulation, creating a simulated real environment can reduce time and improve test efficiency. Using the test data when the vehicle pre-drives in the real environment for testing can ensure the authenticity of the data, thereby improving the accuracy of the test result. The autonomous driving algorithm module can also obtain the result of the vehicle's execution according to the control instruction to complete the closed-loop test, further improving the accuracy of the test result.

[0072] In a possible implementation, the i-th test data played by the data playback module 101 may include obstacle information of the vehicle. For example, the obstacle may be a movable obstacle such as a pedestrian or another vehicle on the road. The i-th test data may store the original relative data of the obstacle relative to the vehicle. For example, the relative position and relative speed between the obstacle and the vehicle. Since in the simulated real environment, the position and speed of the vehicle may not be the same as those of the vehicle in the original test data, it is not accurate to directly apply the original relative data of the obstacle relative to the vehicle in the original test data to the simulated real environment. Based on this, after receiving the i-th test data, when the i-th test data includes a movable obstacle, the data simulation module 102 may determine the true relative data of the obstacle relative to the vehicle in the current simulated real environment based on the original relative data of the obstacle relative to the vehicle, so as to generate more accurate control instructions based on accurate obstacle information and improve the accuracy of testing the autonomous driving algorithm. Optionally, the vehicle can obtain obstacle information, including information such as whether the obstacle is movable, during the process of generating test data when driving in the real environment in advance.

[0073] Specifically, the absolute data of the obstacle can be determined based on the i-th vehicle original state information and the original relative data in the i-th test data. Wherein, the i-th vehicle original state information may include the position and speed of the vehicle, the original relative data may include the relative position and relative speed of the obstacle relative to the vehicle, and the absolute data may include the position and speed of the obstacle. Then, based on the absolute data of the obstacle and the current state information of the vehicle corresponding to the i-th test data, the true relative data of the obstacle relative to the vehicle in the simulated real environment is determined, and the true relative data is fed back to the autonomous driving algorithm module 103. Wherein, the current state information of the vehicle may include the position and speed of the vehicle in the simulated real environment. Therefore, based on the absolute position of the obstacle and the position of the vehicle in the simulated real environment, the true relative position of the obstacle relative to the vehicle in the simulated real environment can be determined, and based on the absolute speed of the obstacle and the speed of the vehicle in the simulated real environment, the relative speed of the obstacle relative to the vehicle in the simulated real environment can be determined, that is, the true relative data of the obstacle relative to the vehicle in the simulated real environment is determined.

[0074] The autonomous driving algorithm module 103 may update the original relative data in the i-th test data with the true relative data, that is, the original relative data can be replaced with the true relative data to obtain the updated i-th test data. Then, the autonomous driving algorithm is executed based on the updated i-th test data and the current state information of the vehicle corresponding to the i-th test data, and the i-th control instruction is output.

[0075] When the test data includes movable obstacles, the data simulation module can correct the obstacle data in the original test data, thereby improving the accuracy of the data input into the autonomous driving algorithm module and enhancing the accuracy of testing the autonomous driving algorithm.

[0076] For details, please refer to Figure 3 shown in Figure 3 which is a schematic structural diagram of another autonomous driving test system provided by an embodiment of the present application.

[0077] In a possible implementation manner, the autonomous driving test system may further include: a data processing module. According to Figure 3 it can be known that the system 100 includes: a data playback module 101, a data simulation module 102, an autonomous driving algorithm module 103, a test module 104, and a data processing module 105.

[0078] The data processing module 105 can obtain the original test data generated when the vehicle travels in advance. Classify and slice the original test data based on test requirements to obtain multiple test data and labels corresponding to each test data. That is, the original test data generated when the vehicle travels in a real environment in advance can be stored in the data processing module 105. The data processing module 105 can divide the original test data into different categories, such as map data, vehicle parameters, movable obstacles, etc., and can also slice the test data according to the duration requirement, so as to divide the original test data into multiple test data, and each test data can save the corresponding label for indexing the corresponding test data based on the label.

[0079] Since each piece of original test data generated by the vehicle has a timestamp, after obtaining multiple test data, each timestamp has corresponding test data. The data processing module 105 can arrange the multiple test data in timestamp order and send the arranged multiple test data to the data playback module 101. So that the data playback module 101 can play the corresponding test data in timestamp order to restore the real environment when the vehicle travels and test the autonomous driving algorithm.

[0080] In a possible implementation manner, when playing the test data, the data playback module 101 can adjust the playback rate of the test data. That is, the data playback module 101 can pre-set playback functions at different speeds and play the test data at different speeds. For example, the test data can be played at 1.25 times speed, 1.5 times speed, 0.75 times speed, 0.5 times speed, etc.

[0081] Optionally, the data playback module 101 can also be set with a pause / resume playback function. During the playback of test data, the pause playback function can be triggered to pause the playback of test data. After the test data is paused, the resume playback function can be triggered to resume the playback of test data.

[0082] Optionally, the data playback module 101 can also be set with a jump playback function. Since there is test data corresponding to multiple timestamps, the data playback module 101 can directly jump to the test data corresponding to a certain timestamp for playback to meet different test requirements.

[0083] For details, please refer to Figure 4 shown in Figure 4 which is a schematic structural diagram of another autonomous driving test system provided by an embodiment of the present application.

[0084] In a possible implementation, the autonomous driving test system may further include: a middleware module. According to Figure 4 it can be known that the system 100 includes: a data playback module 101, a data simulation module 102, an autonomous driving algorithm module 103, a test module 104, a data processing module 105, and a middleware module 106.

[0085] In an actual application scenario, the format of the test data played by the data playback module 101 may be different from the data format input to the data simulation module 102 and the data format input to the autonomous driving algorithm module 103. Therefore, a middleware module 106 can be added to perform different data format conversions on the test data.

[0086] Specifically, after receiving the test data played by the data playback module 101, the middleware module 106 can perform data format conversion on the test data to obtain first converted test data that conforms to the input data format of the autonomous driving algorithm module 103, and send the first converted test data to the autonomous driving algorithm module 103. The middleware module 106 can also perform data format conversion on the test data to obtain second converted test data that conforms to the input data format of the data simulation module 102, and send the second converted test data to the data simulation module 102.

[0087] For details, please refer to Figure 5 shown in Figure 5 which is a schematic structural diagram of another autonomous driving test system provided by an embodiment of the present application.

[0088] In a possible implementation, the autonomous driving test system may further include: a visualization module. According to Figure 5It can be known that the system 100 includes: a data playback module 101, a data simulation module 102, an autonomous driving algorithm module 103, a test module 104, a data processing module 105, a middleware module 106, and a visualization module 107.

[0089] Since the data simulation module 102 can simulate the real environment during vehicle driving, the visualization module 107 can render and display the three-dimensional scene of the vehicle driving in the simulated real environment based on the simulation parameters of the data simulation module 102, enabling testers to intuitively understand the driving state of the vehicle. Based on this, when testing the autonomous driving algorithm based on the vehicle's state information, it is also possible to observe whether the vehicle's behavior meets the requirements in the visualization module 107, thereby facilitating the testing of the autonomous driving algorithm more conveniently.

[0090] According to the above embodiments, when the data playback module 101 plays the test data, it can adjust the playback rate of the test data, pause / resume the playback of the test data, and also jump to play the test data. Based on this, function buttons for controlling the display and playback of the test data can be set in the visualization module 107. For example, the function buttons can include a fast forward button, a slow play button, playback buttons at different speeds, a pause playback button, a resume playback button, a jump playback button, etc. The test data can be controlled to play according to the corresponding functions by clicking the function buttons.

[0091] That is, the visualization module 107 supports user interaction operations on the visualization module 107. When the user clicks the function buttons with different playback functions, the data playback module 101 can play the test data according to the corresponding functions.

[0092] In addition, the user can also adjust different viewing angles for observing the simulated real environment in the visualization module 107. Parameters such as the vehicle's state information can be displayed in the visualization module 107 to facilitate intuitively determining the test results of the autonomous driving algorithm. Additionally, in combination with different test requirements, the intermediate results of the autonomous driving algorithm can be displayed in the visualization module 107, and testers are supported to mark important parameters, etc.

[0093] Based on the above system embodiments, the embodiments of the present application also provide a test method for an autonomous driving algorithm. Specifically, refer to Figure 6 as shown Figure 6 which is a flowchart of a test method for an autonomous driving algorithm provided by the embodiments of the present application.

[0094] Optionally, this method can be applied to the autonomous driving algorithm module, and this method can include the following steps:

[0095] S601: Obtain the i-th test data of the vehicle.

[0096] Among them, the test data is the data generated when the vehicle travels in a real environment in advance. The test data includes n test data corresponding to n timestamps. The i-th test data can represent the test data corresponding to the i-th timestamp, where i = 1, 2, …, n, and n represents the number of timestamps (test data).

[0097] Optionally, the data processing module can obtain the original test data generated when the vehicle travels in advance, classify or slice the original test data to obtain multiple test data. Then, the multiple test data are arranged in timestamp order to obtain the test data corresponding to each timestamp. The data processing module sends the test data to the data playback module in timestamp order. During the test, the data playback module plays the test data in timestamp order, that is, the i-th timestamp corresponds to the i-th test data.

[0098] S602: Obtain the current state information of the vehicle corresponding to the i-th test data determined by the data simulation module.

[0099] Among them, the current state information can include the current position, speed, acceleration, etc. of the vehicle. That is, when the automatic driving algorithm module receives the i-th test data, it can obtain the latest current state information of the vehicle from the data simulation module, so that the automatic driving algorithm module can execute the automatic driving algorithm based on the accurate state information of the vehicle.

[0100] Among them, the data simulation module is used to generate a simulated real environment according to the i-th test data. At the initial test, based on the first original vehicle state information in the first test data, the vehicle can be controlled to travel in the simulated real environment. The first original vehicle state information includes the position, speed, acceleration, etc. of the vehicle corresponding to the first timestamp. When the vehicle travels in the simulated real environment, the data simulation module can determine the state information of the vehicle.

[0101] S603: Based on the i-th test data and the current state information, execute the automatic driving algorithm and output the i-th control instruction. Optionally, the automatic driving algorithm module can send the i-th control instruction to the data simulation module, and the data simulation module controls the vehicle to travel in the simulated real environment based on the i-th control instruction and determines the state information of the vehicle.

[0102] S604: Update i with i + 1 and repeat the above process until i = n, so that the test module can determine the test result of the automatic driving algorithm based on the state information of the vehicle.

[0103] Since there are multiple test data corresponding to multiple timestamps during the test process, the above process is a repeated execution process. That is, when the automatic driving algorithm module receives the (i + 1)-th test data corresponding to the next (i + 1)-th timestamp, it also executes the automatic driving algorithm according to the above steps.

[0104] Optionally, since multiple control functions of the autonomous driving algorithm may need to be tested during the entire test process, the data simulation module controls the vehicle to travel in a simulated real environment based on multiple control instructions, and determines multiple state information of the vehicle after executing the control instructions. Optionally, when the test module determines that any one of the state information does not conform to the preset control strategy, it determines that the test result of the autonomous driving algorithm is unqualified.

[0105] Since the original vehicle state information included in the test data may be different from the state information of the vehicle after executing the control instructions, in the embodiments of the present application, the autonomous driving algorithm module may continue to execute the autonomous driving algorithm based on the latest current state information of the vehicle and the test data, thereby improving the accuracy of testing the autonomous driving algorithm.

[0106] In a possible implementation manner, the autonomous driving algorithm module may be a module in the autonomous driving test system, and the autonomous driving test system may further include: a data playback module, a data simulation module, and a test module. It should be noted that the implementation principles of each module in the autonomous driving test system may refer to the above system embodiments and will not be elaborated here.

[0107] Through the method provided by the embodiments of the present application, the autonomous driving algorithm module can use the test data of the vehicle traveling in the real environment for testing, ensuring the authenticity of the data, thereby improving the accuracy of the test result. The autonomous driving algorithm module can also obtain the latest state information of the vehicle after executing the control instructions, complete the closed-loop test, and further improve the accuracy of the test result.

[0108] Based on the above system embodiments, the embodiments of the present application further provide a test device for an autonomous driving algorithm. Specifically, see Figure 7 as shown in Figure 7 which is a schematic diagram of a test device for an autonomous driving algorithm provided by the embodiments of the present application.

[0109] Optionally, the device may be applied to the autonomous driving algorithm module, and the device 700 may include:

[0110] A first acquisition unit 701, configured to acquire the i-th test data of the vehicle, where the i-th test data is data generated when the vehicle travels in the real environment in advance. The test data includes n test data corresponding to n timestamps, and the i-th test data represents the test data corresponding to the i-th timestamp, i = 1, 2,..., n;

[0111] A second acquisition unit 702, configured to acquire the current state information of the vehicle corresponding to the i-th test data determined by the data simulation module;

[0112] The algorithm execution unit 703 is configured to execute an autonomous driving algorithm based on the i-th test data and the current status information, and output the i-th control instruction; wherein, the data simulation module is configured to generate a simulated real environment according to the i-th test data; control the vehicle to travel in the simulated real environment based on the first original vehicle status information in the first test data; and determine the status information of the vehicle; and, based on the i-th control instruction, control the vehicle to travel in the simulated real environment and determine the status information of the vehicle.

[0113] The iteration unit is configured to update i with i + 1 and repeat the above process until i = n, so that the test module determines the test result of the autonomous driving algorithm based on the status information of the vehicle.

[0114] Based on the above method embodiments and device embodiments, an embodiment of the present application further provides an electronic device. This will be introduced below with reference to the accompanying drawings.

[0115] See Figure 8 , Figure 8 which is a schematic diagram of an electronic device provided by an embodiment of the present application.

[0116] The device 800 includes: a memory 801 and a processor 802;

[0117] The memory 801 is used to store relevant program codes;

[0118] The processor 802 is configured to call the program codes and execute the test method of the autonomous driving algorithm described in the above method embodiment.

[0119] In addition, an embodiment of the present application further provides a computer-readable storage medium, which is used to store a computer program, and the computer program is used to execute the test method of the autonomous driving algorithm described in the above method embodiment.

[0120] An embodiment of the present application further provides a computer program product, which includes computer programs / instructions, and when the computer programs / instructions are executed by a processor, the test method of the autonomous driving algorithm described in the above method embodiment is implemented.

[0121] It should be noted that the above-mentioned computer-readable medium of the present application can be a computer-readable signal medium, a computer-readable storage medium, or any combination of the two. The computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the above. More specific examples of the computer-readable storage medium can include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0122] The computer program product can be written in any combination of one or more programming languages for programming code to perform the operations of the embodiments of the present application. The programming languages include object-oriented programming languages such as Java, C++, etc., and also include conventional procedural programming languages such as the "C" language or similar programming languages. The programming code can be executed entirely on the user computing device, partially on the user device, executed as an independent software package, partially on the user computing device and partially on a remote computing device, or entirely on a remote computing device or server.

[0123] It should be noted that the various embodiments in this specification are described in a progressive manner. Each embodiment focuses on the differences from other embodiments. The same or similar parts among the various embodiments can be referred to each other. In particular, for the system or apparatus embodiments, since they are basically similar to the method embodiments, the description is relatively simple. For the relevant parts, refer to the partial description of the method embodiments. The device embodiments described above are merely illustrative. The units or modules described as separate components may or may not be physically separated. The components shown as units or modules may or may not be physical modules, that is, they can be located in one place or distributed to multiple network units. One can select some or all of the units or modules according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art can understand and implement it without creative efforts.

[0124] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functions, and operations that the methods, apparatuses, devices, etc. according to various embodiments of the present application may implement. In this regard, each block in the flowchart or block diagram may represent a module, a program segment, or a portion of code that contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order than that marked in the accompanying drawings. For example, two consecutive blocks shown may actually be executed substantially in parallel, and they may sometimes be executed in the reverse order, depending on the functions involved. It should also be noted that each block in the block diagram and / or flowchart, as well as combinations of blocks in the block diagram and / or flowchart, may be implemented by a dedicated hardware-based system that performs the specified functions or operations, or may be implemented by a combination of dedicated hardware and computer instructions.

[0125] It should be understood that in the present application, "at least one (item)" means one or more, and "a plurality" means two or more. "And / or" is used to describe the association relationship of associated objects, indicating that there can be three relationships. For example, "A and / or B" can mean: only A exists, only B exists, and both A and B exist at the same time. Among them, A and B can be singular or plural. The character " / " generally indicates that the associated objects before and after are in an "or" relationship. "At least one (one) of the following" or its similar expression refers to any combination of these items, including any combination of single items (ones) or plural items (ones). For example, at least one (one) of a, b, or c can mean: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, and c can be single or multiple.

[0126] It should also be noted that in the present application, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprise", "include" or any other variant thereof are intended to cover non-exclusive inclusion, so that a process, method, article, or device that includes a series of elements not only includes those elements, but also includes other elements not expressly listed, or also includes elements inherent to such process, method, article, or device. Without further limitation, an element defined by the statement "including one..." does not exclude the existence of additional identical elements in the process, method, article, or device that includes the element.

[0127] The steps of the methods or algorithms described in connection with the embodiments disclosed in this application may be implemented directly in hardware, in software modules executed by a processor, or in a combination thereof. The software modules may be located in random access memory (RAM), internal memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, removable disk, CD-ROM, or any other form of storage medium known in the art.

[0128] The foregoing description of the disclosed embodiments enables those skilled in the art to make or use the present application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Thus, the present application is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. An autonomous driving test system, characterized in that, The system includes: a data playback module, a data simulation module, an autonomous driving algorithm module, and a test module; The data playback module is used to play the test data of the vehicle. The test data includes n test data corresponding to n timestamps, and the test data is the data generated by the vehicle during its previous real - environment driving; The data simulation module is used to generate a simulated real environment based on the test data; control the vehicle to drive in the simulated real environment based on the first original vehicle state information in the first test data; and determine the state information of the vehicle; The autonomous driving algorithm module is used to execute the autonomous driving algorithm based on the i - th test data and the current state information of the vehicle obtained from the data simulation module, and output the i - th control instruction. The i - th test data represents the test data corresponding to the i - th timestamp, where i = 1, 2, …, n; The data simulation module is further used to control the vehicle to drive in the simulated real environment based on the i - th control instruction, and determine the state information of the vehicle; Update i with i + 1, and repeat the above process until i = n; The test module is used to determine the test result of the autonomous driving algorithm based on the state information of the vehicle.

2. The system according to claim 1, wherein The data playback module is used to play the corresponding n test data in the order of n timestamps. The i - th test data includes the i - th environmental data and the i - th original vehicle state information; The autonomous driving algorithm module is used to, when receiving the i - th test data corresponding to the i - th timestamp, obtain the current state information of the vehicle from the data simulation module. The current state information is the latest state information of the vehicle in the simulated real environment corresponding to the i - th timestamp; execute the autonomous driving algorithm based on the i - th environmental data and the current state information of the vehicle, and output the i - th control instruction.

3. The system according to claim 1, wherein When the i - th test data includes the original relative data of the obstacle relative to the vehicle, the data simulation module is further used to determine the absolute data of the obstacle based on the i - th original vehicle state information and the original relative data in the i - th test data; determine the real relative data of the obstacle relative to the vehicle in the simulated real environment based on the absolute data and the current state information corresponding to the i - th test data, and feedback the real relative data to the autonomous driving algorithm module; The autonomous driving algorithm module is used to update the original relative data in the i - th test data based on the real relative data, and execute the autonomous driving algorithm based on the updated i - th test data and the current state information corresponding to the i - th test data, and output the i - th control instruction.

4. The system according to claim 1, wherein The test module is used to determine that the test result of the autonomous driving algorithm is qualified when the state information meets the preset control strategy; otherwise, determine that the test result of the autonomous driving algorithm is unqualified.

5. The system according to claim 1, wherein The system further includes: a middleware module, configured to receive the test data; convert the data format of the test data to obtain first-converted test data, and send the first-converted test data to the autonomous driving algorithm module; convert the data format of the test data to obtain second-converted test data, and send the second-converted test data to the data simulation module.

6. The system according to claim 1, characterized in that, The system further includes: a data processing module, configured to obtain the original test data when the vehicle travels in advance; classify and slice the original test data based on test requirements to obtain a plurality of the test data and labels corresponding to the test data; and send the plurality of test data to the data playback module in timestamp order.

7. The system according to claim 1, characterized in that, The system further includes: a visualization module, configured to render and display a three-dimensional scene of the vehicle traveling in the simulated real environment; alternatively, display function buttons for controlling the playback of the test data; or support interactive operations of the user on the visualization module.

8. The system according to any one of claims 1 to 7, characterized in that, The data playback module is configured to adjust the playback rate of the test data; alternatively, pause / resume the playback of the test data; or skip to play the test data.

9. A test method for an autonomous driving algorithm, characterized in that, The method is applied to an autonomous driving algorithm module, and the method includes: Obtain the i-th test data of the vehicle, where the i-th test data is data generated when the vehicle travels in the real environment in advance. The test data includes n test data corresponding to n timestamps, and the i-th test data represents the test data corresponding to the i-th timestamp, i = 1, 2,..., n; Obtain the current state information of the vehicle determined by the data simulation module corresponding to the i-th test data; Based on the i-th test data and the current state information, execute an autonomous driving algorithm and output an i-th control instruction; wherein, the data simulation module is configured to generate a simulated real environment according to the i-th test data; control the vehicle to travel in the simulated real environment based on the first original vehicle state information in the first test data; and determine the state information of the vehicle; and is further configured to control the vehicle to travel in the simulated real environment based on the i-th control instruction and determine the state information of the vehicle; Update i with i + 1, and repeat the above process until i = n, so that the test module determines the test result of the autonomous driving algorithm based on the state information of the vehicle.

10. An electronic device, the device comprising: A memory and a processor; The memory is used to store relevant program codes; The processor is used to call the program codes to execute the test method of the autonomous driving algorithm according to claim 9.

Images