Virtual-real combined automatic driving vehicle lane changing function test system
Through the lane-changing function test system of the self-driving vehicle combining virtual and real, virtual traffic scenarios are built, driving tasks are generated and performance evaluation is evaluated, and the problems of high testing costs and inaccurate test results are solved in the existing technology, achieving high efficiency in multiple scenarios , accurate lane change performance evaluation.
Patent Information
- Application Number
- CN202411147080.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-21
- Publication Date
- 2024-11-26
- Estimated Expiration
- 2044-08-21
AI Technical Summary
In the prior art, the lane-changing function test of autonomous driving vehicles is expensive and cannot be accurately tested in multiple scenarios. Virtual simulation methods are difficult to simulate human driving behavior, resulting in inaccurate test results.
Using a combination of virtual and real autonomous vehicle lane change function test system, a virtual traffic scene is constructed through a scene generation unit, and an adaptive scene task generator generates multiple driving tasks. The human-computer interaction terminal provides driving guidance information, and the vehicle data acquisition module collects vehicle movements. Information, the test evaluation unit evaluates the lane change performance.
It reduces testing costs, improves testing efficiency and accuracy, and can accurately evaluate the lane change performance of autonomous driving vehicles in multiple scenarios, simulating driver behavior in real traffic environments.
Smart Images

Figure CN119023291A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of autonomous driving testing, and in particular to a virtual-real combined autonomous driving vehicle lane changing function testing system. Background Art
[0002] With the advancement of autonomous driving technology, autonomous driving vehicles will inevitably go through a mixed traffic flow stage consisting of autonomous driving vehicles and manually driven vehicles in the process of being fully put into practical application. Since human driving behavior is highly unpredictable and non-standardized in actual traffic scenarios, autonomous driving vehicles need to maintain good coordination with manually driven vehicles in mixed traffic flow scenarios to achieve a safer and more efficient traffic system. In order to achieve an ideal traffic system, verification and testing of various functions of autonomous driving vehicles in mixed traffic flows are essential. Among them, the lane changing function is a high-risk operation faced by autonomous driving vehicles in complex traffic environments, and must pass rigorous testing to ensure safety, reliability and efficiency on real roads.
[0003] In related technologies, in order to verify the lane-changing function of an autonomous vehicle in mixed traffic flow, if the traditional autonomous vehicle lane-changing function test method based on real roads is used, it will take a lot of time and manpower costs, and it is limited by real road traffic conditions and environmental conditions, and it is impossible to test the lane-changing performance of unmanned vehicles in multiple scenarios. If a fully virtual simulation autonomous vehicle lane-changing function test method is used, although it can save a lot of time and cost and simulate various road traffic scenarios, it is still difficult to simulate the complex driving behavior of human drivers, resulting in the test scenario not being close enough to the real scenario, making the test results inaccurate.
[0004] Therefore, the testing scheme for the lane changing performance of autonomous driving vehicles provided in the relevant technology still has room for improvement. Summary of the invention
[0005] The main purpose of the present invention is to provide a virtual-real combined autonomous driving vehicle lane changing function test system, aiming to solve the technical problem of high cost required for lane changing function testing of autonomous driving vehicles in related technologies.
[0006] The present application provides a virtual-real combined autonomous driving vehicle lane-changing function test system, comprising: a scenario generation unit, an adaptive scenario task generator, a human-computer interaction terminal, a vehicle data acquisition module, a test evaluation unit, and a plurality of driving simulators;
[0007] The scenario generating unit is used to construct a virtual traffic scenario according to the input scenario building data, wherein the virtual traffic scenario is used to test the autonomous driving vehicle;
[0008] The adaptive scenario task generator is used to deconstruct the virtual traffic scenario to generate a plurality of different driving tasks, each of which corresponds to a driving vehicle;
[0009] The human-computer interaction terminal is used to generate corresponding driving guidance information according to the driving task, and the driving guidance information is used to guide the corresponding vehicle to operate in the virtual traffic scene;
[0010] The driving simulator is used to perform corresponding driving operations according to the driving guidance information;
[0011] The vehicle data acquisition module is used to collect vehicle motion information, vehicle control information, surrounding vehicle information and traffic scene information of the autonomous driving vehicle and the multiple driving simulators during the test;
[0012] The test evaluation unit is used to evaluate the lane changing performance of the autonomous driving vehicle during the autonomous driving process based on the collected vehicle motion information, vehicle control information, surrounding vehicle information and traffic scene information of the autonomous driving vehicle and the multiple driving simulators during the test process.
[0013] In a possible design, the input scene construction data includes one or more of real scene data, data processed based on real scene data, and customized manually constructed data.
[0014] In a possible design, the human-computer interaction terminal includes a display; the display is used to display the driving guidance information in real time.
[0015] In a possible design, the display includes a first display area, a second display area, and a third display area;
[0016] The first display area is a first viewing angle corresponding to the driving simulator, and the first display area is used to display an environment image around the current driving vehicle in the form of an image stream;
[0017] The second display area is used to display the vehicle motion information, vehicle control information and environmental information of the current vehicle during driving; the motion information includes the current vehicle speed value and the current coordinates of the vehicle, the vehicle control information includes the vehicle throttle amount, brake pedal amount and steering wheel angle; the environmental information includes the current scene name, the vehicles around the vehicle and the distance between the surrounding vehicles and the vehicle;
[0018] The third display area is used to display the driving guidance information, and the driving guidance information is used to guide the current vehicle to complete the driving task in the virtual traffic scene.
[0019] In one possible design, the evaluating the lane changing performance of the autonomous driving vehicle during the autonomous driving process includes:
[0020] Setting a plurality of lane changing performance indicators for evaluating lane changing performance, wherein the plurality of lane changing performance indicators include lane changing success rate, minimum safety distance, lane changing time, and passenger comfort during lane changing;
[0021] Determining score values of the multiple lane change performance indicators according to vehicle motion information, vehicle control information, surrounding vehicle information, and traffic scene information of the autonomous driving vehicle and the multiple driving simulators during the test;
[0022] The lane changing performance of the autonomous driving vehicle during the autonomous driving process is determined according to the score values of the multiple lane changing performance indicators.
[0023] In one possible design, the evaluating the lane changing performance of the autonomous driving vehicle during the autonomous driving process further includes:
[0024] Determine a weight value corresponding to each lane changing performance indicator;
[0025] According to the score values of the multiple lane changing performance indicators and the weight values corresponding to the respective lane changing performance indicators, a weighted sum is performed on the respective lane changing performance indicators to obtain a total score value of the lane changing performance; the total score value of the lane changing performance is used to characterize the lane changing performance of the autonomous driving vehicle during the autonomous driving process.
[0026] In one possible design, the adaptive scenario task generator includes a scenario deconstruction module, a driving task generation module, and a driving task allocation module;
[0027] The scene deconstruction module is used to deconstruct the virtual traffic scene to obtain scene information for lane change testing;
[0028] The driving task generating module is used to generate a plurality of different driving tasks according to the scenario information of the lane change test;
[0029] The driving task allocation module is used to allocate the multiple different driving tasks to corresponding driving simulators.
[0030] In a possible design, the test evaluation unit is further used to: evaluate the performance of the manually driven vehicle during the test according to the vehicle motion information, vehicle control information, surrounding vehicle information and traffic scene information of each driving simulator during the test;
[0031] Among them, the evaluation indicators for evaluating the performance of manually driven vehicles during the test process include one or more of the task completion rate, task completion time, average speed during driving, lane keeping stability, smoothness of steering wheel control, number of collisions during task execution, and emergency braking reaction time.
[0032] In a possible design, the evaluation index of lane keeping stability includes one or more of lane departure frequency and amplitude, lane position consistency, and steering control accuracy.
[0033] The present application provides a lane-changing function test system for an autonomous driving vehicle, which uses a scenario generation unit to construct a virtual traffic scenario based on input scenario construction data. At the same time, an adaptive scenario task generator deconstructs the virtual traffic scenario to generate multiple different driving tasks to simulate the dynamic behavior of the driver in a real traffic environment. The vehicle data acquisition module collects relevant information about the autonomous driving vehicle and the manually driven vehicle during the test. Finally, the test evaluation unit evaluates the lane-changing performance of the autonomous driving vehicle during the autonomous driving process based on the collected relevant information about the autonomous driving vehicle and the manually driven vehicle. The test system provided by the present application overcomes the technical problem of high test cost in the related art when testing the lane-changing performance of unmanned vehicles; at the same time, different scenario information can be input in the present application to construct different virtual test environments, so as to realize the testing of the lane-changing performance of autonomous driving vehicles in multiple scenarios, thereby improving the test efficiency and test accuracy. BRIEF DESCRIPTION OF THE DRAWINGS
[0034] Figure 1 A schematic diagram of the structure of a lane-changing function test system for an autonomous driving vehicle provided in an embodiment of the present application;
[0035] Figure 2 A schematic diagram of the structure of an adaptive scenario task generator provided in an embodiment of the present application;
[0036] Figure 3 A schematic diagram of multiple functional display areas on a display provided in an embodiment of the present application;
[0037] Figure 4 A schematic diagram of a test evaluation process of a test system is provided for an embodiment of the present application;
[0038] Figure 5 Schematic diagram of the lane changing performance evaluation method provided in an embodiment of the present application.
[0039] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION
[0040] The technical solution in this application will be described below in conjunction with the accompanying drawings.
[0041] In the description of the embodiments of the present application, unless otherwise specified, “ / ” means or. For example, A / B can mean A or B. The “and / or” in this article is merely a way to describe the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B can mean: A exists alone, A and B exist at the same time, and B exists alone.
[0042] In the following, the terms "first" and "second" are used for descriptive purposes only and are not to be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features. In the description of this embodiment, unless otherwise specified, "plurality" means two or more.
[0043] In the related art, when testing the lane-changing function of an autonomous vehicle, one test scheme is to use a traditional lane-changing function test method for autonomous vehicles based on real roads. This test method requires a lot of time and manpower costs, and due to the limitations of real road traffic and real environmental conditions, the real scenes that can be tested are somewhat one-sided, and it is impossible to fully realize the repeated and safe lane-changing performance test of autonomous vehicles in multiple scenes. Another test scheme is to use a completely virtual simulation lane-changing function test method for autonomous vehicles. Although it can save a lot of time and simulate various road traffic scenes, this method has problems such as difficulty in simulating the complex behavior of human drivers, and because the behavior of drivers in mixed traffic flows is difficult to quantify and reproduce, the test results of autonomous vehicles are not accurate enough.
[0044] In order to overcome the defects in the related art, the present application provides a virtual and real combination of autonomous driving vehicle lane changing function test system, which uses a scene generation unit to construct a virtual traffic scene according to the input scene construction data. At the same time, the adaptive scene task generator deconstructs the virtual traffic scene to generate multiple different driving tasks to test the manually driven vehicle. The vehicle data acquisition module collects the relevant information of the autonomous driving vehicle and the manually driven vehicle during the test. Finally, the test evaluation unit evaluates the lane changing performance of the autonomous driving vehicle during the autonomous driving process based on the collected relevant information of the autonomous driving vehicle and the manually driven vehicle. It can be seen that according to the test system of the present application, a variety of different test scenarios can be simulated by changing the input scene construction data. At the same time, the human-computer interaction terminal and the driving simulator can be used to simulate the impact of the manually driven vehicle on the autonomous driving vehicle when performing the lane changing task in the actual situation, thereby saving the test cost and improving the test accuracy.
[0045] Figure 1For a schematic diagram of the structure of the lane changing function test system for an autonomous driving vehicle provided in the embodiment of the present application, please refer to Figure 1 As shown, the autonomous driving vehicle lane changing function test system includes: a scenario generation unit 10, an adaptive scenario task generator 20, a human-computer interaction terminal 30, a vehicle data acquisition module 40, a test evaluation unit 50 and multiple driving simulators.
[0046] Among them, the scene generation unit 10 is used to construct a virtual traffic scene according to the input scene construction data, and the virtual traffic scene is used to test the autonomous driving vehicle. The adaptive scene task generator 20 is used to deconstruct the virtual traffic scene to generate multiple different driving tasks, each driving task corresponds to a driving simulator, that is, one driving task corresponds to a manually driven vehicle. The human-computer interaction terminal 30 is used to generate corresponding driving guidance information according to the driving task, and the driving guidance information is used to guide the corresponding manually driven vehicle to operate in the virtual traffic scene. The driving simulator is used to perform corresponding driving operations according to the driving guidance information to simulate the behavior of the manually driven vehicle. The vehicle data acquisition module 40 is used to collect vehicle motion information, vehicle control information, surrounding vehicle information and traffic scene information of the autonomous driving vehicle and multiple driving simulators during the test process. The test evaluation unit 50 is used to evaluate the lane changing performance of the autonomous driving vehicle during the autonomous driving process based on the collected vehicle motion information, vehicle control information, surrounding vehicle information and traffic scene information of the autonomous driving vehicle and multiple driving simulators during the test process.
[0047] It can be understood that the multiple driving simulators in the present application are equivalent to simulating multiple manually driven vehicles in real traffic scenarios, and each driver controls a driving simulator according to a driving guidance information to simulate the scenario of manually driving a vehicle in a real scenario.
[0048] The vehicle data acquisition module 40 in the present application may include multiple sensors and data processing chips. The multiple sensors may collect different raw data. The data processing chip processes the collected raw data to obtain vehicle motion information, vehicle control information, surrounding vehicle information, and traffic scene information. Different information may be collected from different modules or hardware. For example, vehicle control information may be collected from a driving simulator.
[0049] In some embodiments, the scene construction data input into the scene generation unit 10 includes one or more of real scene data, data processed based on the real scene data, and customized manually constructed data.
[0050] The data processed based on the real scene data may be data obtained by enhancing the real scene data or generalizing the real scene data.
[0051] The custom data may be some test data set by the tester according to the test requirements, so as to enhance the complexity of the virtual test environment or increase the diversity of the virtual test environment.
[0052] In this embodiment, different virtual traffic scenes can be constructed based on one or more of real scene data, data processed based on real scene data, and customized manually constructed data. In other words, virtual traffic scenes of different complexities can be constructed.
[0053] It can be understood that the adaptive scenario task generator 20 generates a virtual traffic scene based on the constructed virtual traffic scene for deconstruction to generate multiple different driving tasks, and each driving simulator performs different driving tasks to simulate the situation of manually driving a vehicle in a real scene, so as to improve the authenticity and accuracy of the test.
[0054] Figure 2 For a schematic diagram of the structure of the adaptive scenario task generator provided in the embodiment of the present application, see Figure 2 As shown, the adaptive scenario task generator 20 includes a scenario deconstruction module 201 , a driving task generation module 202 and a driving task allocation module 203 .
[0055] The scene deconstruction module 201 is used to deconstruct the virtual traffic scene to obtain scene information for lane change test. The driving task generation module 202 is used to generate multiple different driving tasks according to the scene information of lane change test; the driving task allocation module 203 is used to allocate multiple different driving tasks to corresponding driving simulators.
[0056] The adaptive scenario task generator 20 of this embodiment is used to deconstruct the scenario generated by the scenario generation unit 10, generate different driving tasks after deconstruction and output them to different driving simulators, so that different driving simulators are in the same traffic scenario and environment while each driver has different driving tasks and goals, ultimately achieving the coordinated operation of autonomous driving vehicles and manually driven vehicles in mixed traffic flows.
[0057] It can be understood that the adaptive scenario task generator 20 generates different driving tasks and goals for each virtual traffic scene by writing script files, wherein each driving task is in the form of text and images as instructions or guidance, and the generated driving tasks and goals are injected into the human-computer interaction terminal 30 of each driving simulator.
[0058] The human-computer interaction terminal 30 of this embodiment includes a display; the display is used to display driving guidance information in real time and provide the driver with instruction information for executing the driving simulator.
[0059] The human-computer interaction terminal 30 in this embodiment is used to provide each driver with a picture of the current vehicle running in a virtual scene when operating the driving simulator, and to display in real time on the large screen of the driving simulator the vehicle-related motion data, current position information, surrounding vehicle information and relevant data of the current traffic scene and environment of each driver in the process of performing the driving task. At the same time, the driving task generated by the adaptive scenario task generator 20 is displayed to each driver in real time in the form of images or text to generate route instructions or task guidance, so as to guide them to complete the task of simulating manual driving.
[0060] In some embodiments, the human-computer interaction terminal may further include an input device, such as a keyboard, a mouse, a stylus, etc., so that a technician can modify part of the indication information to meet the diversity of the test.
[0061] In some embodiments, Figure 3 For a schematic diagram of multiple functional display areas on a display provided in an embodiment of the present application, see Figure 3 As shown, the display of this embodiment includes a first display area, a second display area and a third display area.
[0062] The first display area is a first viewing angle corresponding to the driving simulator, and the first display area is used to display the environment image around the current driving vehicle in the form of an image stream. The image stream of this embodiment can be in the form of a picture stream or a video stream, which is not limited here.
[0063] The second display area is used to display the vehicle motion information, vehicle control information and environmental information of the current vehicle during driving. The motion information includes the current vehicle speed value and the current vehicle coordinates (i.e. the vehicle position information), the vehicle control information includes the vehicle throttle amount, brake pedal amount and steering wheel angle, etc.; the environmental information includes the current scene name, the vehicle information around the vehicle, and the distance information between the surrounding vehicles and the vehicle.
[0064] The third display area is used to display driving guidance information for the current vehicle's driving task. The driving guidance information can be understood as instruction information or guidance information. The driving guidance information is used to guide the driver to perform corresponding operations on the driving simulator to control the driving simulator to complete the driving task normally and accurately in the virtual scene and reduce improper operations during driving.
[0065] Among them, this embodiment includes multiple driving simulators, such as Figure 1The test system shown includes a driving simulator 1, a driving simulator 2, etc. The driving simulator is used to provide each driver with a realistic virtual traffic scene and driving experience, which includes the control of the vehicle throttle, brake, steering wheel, gear, headlight, etc. At the same time, the vehicle data acquisition module 40 can obtain the vehicle-related data and current position information of each driver during the driving task, the operating data of surrounding vehicles, and the relevant data of the current traffic scene and environment.
[0066] Figure 4 The test evaluation process diagram of the test system provided in this application embodiment is shown in Figure 4 As shown, the vehicle data acquisition module 40 feeds back the collected relevant information of the manually driven vehicle and the autonomous driving vehicle to the scene generation unit 10, and the scene generation unit 10 then sends the collected relevant information of the vehicle to the test evaluation unit 5.
[0067] The test evaluation unit 50 in this embodiment is used to evaluate the performance of the autonomous driving vehicle in the lane change function test in the virtual scene and to evaluate the performance of the driver when performing the driving task. It uses the vehicle data acquisition module 40 to obtain the vehicle-related data of the autonomous driving vehicle and the manually driven vehicle, and analyzes and processes the data to evaluate the performance of the autonomous driving vehicle and the driver according to different performance indicators.
[0068] Figure 5 For a schematic diagram of the lane changing performance evaluation method provided in the embodiment of the present application, please refer to Figure 5 As shown, in one embodiment, the evaluation of the lane changing performance of the autonomous driving vehicle during the autonomous driving process specifically includes:
[0069] S501. Set a plurality of lane changing performance indicators for evaluating lane changing performance, wherein the plurality of lane changing performance indicators include a lane changing success rate, a minimum safe distance, a lane changing time, and a passenger comfort level during lane changing.
[0070] S502. Determine score values of multiple lane changing performance indicators based on vehicle motion information, vehicle control information, surrounding vehicle information, and traffic scene information of the autonomous driving vehicle and the multiple driving simulators during the test.
[0071] S503: Determine the lane changing performance of the autonomous driving vehicle during the autonomous driving process according to the score values of the multiple lane changing performance indicators.
[0072] Among them, the lane change success rate refers to the proportion of autonomous driving vehicles that successfully complete lane changes when performing lane change operations. Specifically, it indicates how many times the lane change can be completed safely and smoothly among all attempted lane change operations.
[0073] The formula for calculating the lane change success rate is:
[0074] In formula (1), C success Indicates the number of successful lane changes. The criteria for success include: no collision, no emergency braking or avoidance of other vehicles; C total Indicates the total number of lane change attempts.
[0075] By counting the results of each lane change test, the lane change success rate can be calculated. Taking into account the road type, environmental factors and safety of the current scene, a target value for the lane change success rate can be set. For example, if the current test environment is a city road on a rainy day, the target value is usually set between 95% and 98%. The calculated lane change success rate is compared with the target value to determine whether it is within a reasonable range. A scoring mechanism is set based on its performance to obtain a score for the lane change success rate. R_success ,For example:
[0076] Excellent (9-10 points): Lane change success rate ≥ target value + 1%;
[0077] Good (7-8 points): target value ≤ lane change success rate < target value + 1%;
[0078] Average (5-6 points): The lane change success rate is close to but does not reach the benchmark value (target value - 2% ≤ lane change success rate < target value);
[0079] Failure (<5 points): Lane change success rate < target value - 2%.
[0080] The minimum safe distance refers to the minimum distance between the autonomous vehicle and other surrounding vehicles during the lane change process, which is mainly divided into the following situations:
[0081] 1. The distance between the autonomous vehicle and the nearest vehicle in the current lane
[0082] During the lane change process, obtain the minimum distance between the autonomous driving vehicle and the nearest vehicle in front of the current lane.
[0083] The specific calculation formula is: current_front (t)=|P self (t)-P current_front (t)|(2)
[0084] In formula (2), P self (t) represents the position of the autonomous vehicle at time t, P current_front (t) represents the position of the nearest vehicle in front of the current lane at time t.
[0085] 2. The distance between the autonomous vehicle and the closest vehicle behind the target lane
[0086] During the lane change process, obtain the minimum distance between the autonomous driving vehicle and the nearest vehicle behind the target lane.
[0087] The specific calculation formula is: target_rear (t)=|P self (t)-P target_rear (t)|(3)
[0088] Among them, in formula (3), P self (t) represents the position of the autonomous vehicle at time t, P target_rear (t) represents the position of the front vehicle at time t.
[0089] 3. The distance between the autonomous vehicle and the nearest vehicle in the target lane
[0090] During the lane change process, obtain the distance between the autonomous driving vehicle and the nearest vehicle in front of the target lane.
[0091] The specific calculation formula is: target_front (t)=|P self (t)-P target_front (t)|(4)
[0092] In formula (4), P self (t) represents the position of the autonomous vehicle at time t, P target_front (t) represents the position of the front vehicle at time t.
[0093] Combining the minimum distances in all directions mentioned above, the minimum safety distance can be defined as:
[0094] d min =min{d current_front ,d target_rear ,d target_front} (5)
[0095] By collecting the distance between the autonomous driving vehicle and the surrounding vehicles during the lane change process, the minimum safe distance can be calculated. Taking into account the road type, environmental factors and the current vehicle speed in the current scene, an actual benchmark value is set for the minimum safe distance. For example, if the current test environment is an urban road on a rainy day, the benchmark value is usually set to 3-5 meters. The calculated minimum safe distance is compared with the benchmark value to determine whether it is within a reasonable range, and a scoring mechanism is set based on its performance to obtain a score for the minimum safe distance. d_min ,For example:
[0096] Excellent (9-10 points): minimum safety distance ≥ benchmark value + 10%;
[0097] Good (7-8 points): benchmark value ≤ minimum safety distance < benchmark value + 10%;
[0098] Average (5-6 points): The minimum safety distance is close to but does not reach the benchmark value (90%-100% benchmark value);
[0099] Unsafe (<5 points): The minimum safety distance is < 90% of the benchmark value.
[0100] Lane change time refers to the time it takes for an autonomous vehicle to deviate from the current lane and enter the target lane. The calculation formula is:
[0101] T lane-change =t end -t start (6)
[0102] In formula (6), t end Indicates the start time of lane change, usually the moment when the center point of the vehicle begins to deviate from the center line of the current lane; t start Indicates the lane change end time, which is usually the moment when the center of the vehicle completely enters the center line of the target lane and stabilizes in the target lane.
[0103] By recording the start time and end time of the lane changing of the autonomous driving vehicle during the lane changing process, the lane changing time can be calculated.
[0104] Considering the road type, environmental factors, traffic density and current vehicle speed of the current scene, a reasonable time benchmark value is set for the lane change time. For example, if the current test environment is a low-density traffic urban road on a rainy day, the time benchmark value is usually set to 3-5 seconds. The calculated lane change time is compared with the benchmark value to determine whether it is within a reasonable range, and a scoring mechanism is set based on its performance to obtain a score for the lane change time. T_lane-change ,For example:
[0105] Excellent (9-10 points): The lane change time is close to or slightly lower than the benchmark time, and there are no safety hazards;
[0106] Good (7-8 points): The lane change time is slightly longer than the benchmark time, but still within a reasonable range, with no obvious safety hazards;
[0107] Average (5-6 points): The lane change time is significantly longer than the benchmark time and may need to be optimized, but it is still acceptable;
[0108] Unsafe (<5 points): The lane change time is too long or too short, which may pose a safety hazard or a smoothness problem.
[0109] Comfort mainly refers to the smoothness and comfort experienced by passengers during lane changes. Comfort assessment takes into account multiple aspects, including but not limited to acceleration changes, lateral and longitudinal forces, and vehicle dynamic behavior.
[0110] In this embodiment, the acceleration change rate is used as an example to quantitatively calculate the comfort, and the calculation formula is:
[0111] J(t)=da(t) / dt (7)
[0112] Where a(t) represents the acceleration of the autonomous vehicle at time t, and dt represents the time used in the lane change process. The calculated J(t) represents the acceleration change rate of the autonomous vehicle at time t, and a smaller change rate usually means higher comfort.
[0113] By collecting and analyzing the acceleration of the autonomous driving vehicle during the lane changing process as well as the lane changing start time and lane changing end time, the acceleration change rate of the vehicle during the lane changing process can be obtained, thereby obtaining the peak value of the acceleration change rate of the vehicle during the lane changing process.
[0114] The current industry standard for passenger cars considers the peak acceleration rate to be 2-3 m / s. 3 The following is more appropriate, and passengers may feel obvious discomfort if the acceleration rate exceeds this range. Compare the calculated peak value of the acceleration change rate with the above industry standards to determine whether it is within a reasonable range, and set a scoring mechanism based on its performance to obtain a score for comfort. J ,For example:
[0115] Excellent (9-10 points): If the peak value of the acceleration rate of change is less than or equal to 2m / s 3 , the lane change process is very smooth, with almost no noticeable acceleration change, and the passengers feel extremely comfortable;
[0116] Good (7-8 points): If the peak value of the acceleration rate of change is 2-3m / s 3 Within the range, the lane change process is smooth. Although there is a slight change in acceleration, passengers generally do not feel uncomfortable and the comfort is still good.
[0117] General (5-6 points): If the peak value of the acceleration rate is 3-4 m / s 3 If the vehicle is within the specified range, there will be a noticeable change in acceleration during the lane change process, and some passengers may feel a little uncomfortable, but overall it is still within an acceptable range.
[0118] Unsafe (<5 points): If the peak acceleration rate is greater than 4 m / s 3 , the acceleration changes during the lane changing process are large, the passengers will obviously feel uncomfortable and the comfort is poor, and the lane changing strategy or suspension system may need to be optimized.
[0119] In some embodiments, evaluating the lane changing performance of the autonomous driving vehicle during the autonomous driving process specifically also includes: determining a weight value corresponding to each lane changing performance indicator; taking a weighted sum of each lane changing performance indicator according to the score values of the multiple lane changing performance indicators and the weight values corresponding to each lane changing performance indicator to obtain a total score value of the lane changing performance; the total score value of the lane changing performance is used to characterize the lane changing performance of the autonomous driving vehicle during the autonomous driving process.
[0120] In order to comprehensively evaluate the lane changing performance of the autonomous driving vehicle, this embodiment may adopt a more comprehensive evaluation index, and perform a weighted summation of the above performance indicators, which may be expressed as the following formula, for example:
[0121] score=w1score R_success +w2score d_min +w3score T_lane-change +w4score J (8)
[0122] Among them, w1, w2, w3, and w4 are the weights of each indicator, which can be adjusted accordingly according to the specific application scenario.
[0123] According to the pre-set standard, the final calculated score is compared with the standard to evaluate the lane change performance. For example, you can set:
[0124] Excellent (9-10 points), good (7-8 points), fair (5-6 points), poor (<5 points).
[0125] Through the above autonomous driving vehicle motion data collection and performance index calculation, the performance of the autonomous driving vehicle's lane changing function in mixed traffic flow can be comprehensively evaluated. Among them, the specific scoring rules and weight settings can be set according to the actual scenario, which will not be repeated here.
[0126] In some embodiments, the test evaluation unit 50 is also used to evaluate the performance of the manually driven vehicle during the test based on the vehicle motion information, vehicle control information, surrounding vehicle information and traffic scene information of each driving simulator during the test.
[0127] Among them, the evaluation indicators for evaluating the performance of manually driven vehicles during the test process include one or more of the task completion rate, task completion time, average speed during driving, lane keeping stability, smoothness of steering wheel control, number of collisions during task execution, and emergency braking reaction time.
[0128] Among them, the evaluation indicators of lane keeping stability include one or more of lane departure frequency and amplitude, lane position consistency and steering control accuracy.
[0129] It is understandable that there are also multiple different performance indicators for evaluating manually driven vehicles in the performance of driving tasks, such as task completion rate, task completion time, average speed during driving, lane keeping stability, smoothness of steering wheel control, number of collisions and emergency braking reaction time when performing tasks, etc. By analyzing the collected data and calculating the values of different performance indicators, a more comprehensive evaluation of the performance of manually driven vehicles in the performance of driving tasks can be achieved.
[0130] The following is a detailed introduction to the above different performance indicators:
[0131] The task completion rate indicates the pass rate of a driver when performing a series of driving tasks. Each driver's driving task has many different task instructions or guidance. The ratio of the number of completed task instructions or guidance in the entire driving task to the total number of task instructions or guidance is the task completion rate. If this indicator is too low, it means that the generated driving task is too difficult and this scenario will hardly occur in reality; or it means that the driver's driving ability is insufficient and it is necessary to consider replacing the driver for subsequent testing.
[0132] The mission completion time refers to the time it takes for the driver to drive the vehicle to the destination after starting the vehicle after accepting the driving order.
[0133] The average speed during driving indicates the average speed at which the driver controls the vehicle while performing the driving task. This indicator can reflect the complexity of the current lane change test scenario and the driver's proficiency in vehicle control.
[0134] Lane keeping stability refers to the driver's ability and performance to keep the vehicle in the designated lane during driving tasks, including the following evaluation indicators:
[0135] 1. Lane departure frequency and magnitude, usually measured by the number of times the vehicle deviates from the lane centerline and the distance it deviates;
[0136] 2. Lane position consistency, usually measured by the standard deviation of the vehicle's position change relative to the lane centerline;
[0137] 3. Steering control accuracy, usually measured by vehicle steering angle and vehicle trajectory.
[0138] Among them, the smoothness of steering wheel control refers to the driver's ability to control the direction in a smooth, continuous and precise manner when performing driving tasks. Highly smooth steering wheel control usually indicates that the driver has a high level of vehicle control proficiency.
[0139] It can be seen that based on the test system provided by the present application, the scene generation unit 10 realizes the construction of lane change test scenes, traversal of typical traffic scenes and environmental conditions, and other problems, overcoming the technical problems of high cost, low efficiency and poor safety of lane change tests of autonomous driving vehicles on real roads in mixed traffic flows; based on the adaptive scene task generator 20 provided by the present application, it realizes the generation of different driving tasks for each driver in a variety of lane change scene types, solving the technical problem that the driver's behavior in mixed traffic flows is difficult to quantify and reproduce; the test evaluation unit 50 is used to realize the independent evaluation of autonomous driving vehicles and manually driven vehicles in mixed traffic flows. Therefore, the test system of the present application can provide a more comprehensive lane change scene and evaluation for the lane change test function of autonomous driving vehicles in mixed traffic flows, and improve the efficiency, economic benefits and safety of lane change function testing.
[0140] The above embodiments can be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded or executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from a website site, computer, server or data center to another website site, computer, server or data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) mode. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that contains one or more available media sets. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium. The semiconductor medium can be a solid state drive (SSD).
[0141] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the embodiments of the present application.
[0142] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.
[0143] In the several embodiments provided in the embodiments of 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, such as 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.
[0144] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, and may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
[0145] 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.
[0146] If the function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the embodiment of the present application is essentially or the part that contributes to the prior art or the part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions to enable a memory (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (Read-Only Memory, ROM), random access memory (Random Access Memory, RAM), disk or optical disk and other media that can store program code.
[0147] The above is only a specific implementation of the embodiment of the present application, but the protection scope of the embodiment of the present application is not limited thereto. Any technician familiar with the technical field can easily think of changes or replacements within the technical scope disclosed in the embodiment of the present application, which should be included in the protection scope of the embodiment of the present application. Therefore, the protection scope of the embodiment of the present application should be based on the protection scope of the claims.
Claims
1. A virtual and real combined automatic driving vehicle lane changing function test system, characterized in that: include: Scenario generation unit, adaptive scenario task generator, human-computer interaction terminal, vehicle data acquisition module, test and evaluation unit, and multiple driving simulators; The scenario generating unit is used to construct a virtual traffic scenario according to the input scenario building data, wherein the virtual traffic scenario is used to test the autonomous driving vehicle; The adaptive scenario task generator is used to deconstruct the virtual traffic scenario to generate a plurality of different driving tasks, each of which corresponds to a driving vehicle; The human-computer interaction terminal is used to generate corresponding driving guidance information according to the driving task, and the driving guidance information is used to guide the corresponding vehicle to operate in the virtual traffic scene; The driving simulator is used to perform corresponding driving operations according to the driving guidance information; The vehicle data acquisition module is used to collect vehicle motion information, vehicle control information, surrounding vehicle information and traffic scene information of the autonomous driving vehicle and the multiple driving simulators during the test; The test evaluation unit is used to evaluate the lane changing performance of the autonomous driving vehicle during the autonomous driving process based on the collected vehicle motion information, vehicle control information, surrounding vehicle information and traffic scene information of the autonomous driving vehicle and the multiple driving simulators during the test process.
2. The test system according to claim 1, characterized in that: The input scene construction data includes one or more of real scene data, data processed based on real scene data, and customized manually constructed data.
3. The test system according to claim 1 or 2, characterized in that: The human-computer interaction terminal includes a display; the display is used to display the driving guidance information in real time.
4. The test system according to claim 3, characterized in that: The display includes a first display area, a second display area and a third display area; The first display area is a first viewing angle corresponding to the driving simulator, and the first display area is used to display an environment image around the current driving vehicle in the form of an image stream; The second display area is used to display the vehicle motion information, vehicle control information and environmental information of the current vehicle during driving; the motion information includes the current vehicle speed value and the current coordinates of the vehicle, the vehicle control information includes the vehicle throttle amount, brake pedal amount and steering wheel angle; the environmental information includes the current scene name, the vehicles around the vehicle and the distance between the surrounding vehicles and the vehicle; The third display area is used to display the driving guidance information, and the driving guidance information is used to guide the current vehicle to complete the driving task in the virtual traffic scene.
5. The test system according to claim 1 or 2, characterized in that: The evaluating the lane changing performance of the autonomous driving vehicle during the autonomous driving process includes: Setting a plurality of lane changing performance indicators for evaluating lane changing performance, wherein the plurality of lane changing performance indicators include lane changing success rate, minimum safety distance, lane changing time, and passenger comfort during lane changing; Determining score values of the multiple lane change performance indicators according to vehicle motion information, vehicle control information, surrounding vehicle information, and traffic scene information of the autonomous driving vehicle and the multiple driving simulators during the test; The lane changing performance of the autonomous driving vehicle during the autonomous driving process is determined according to the score values of the multiple lane changing performance indicators.
6. The test system according to claim 5, characterized in that: The evaluating the lane changing performance of the autonomous driving vehicle during the autonomous driving process further includes: Determine a weight value corresponding to each lane changing performance indicator; According to the score values of the multiple lane changing performance indicators and the weight values corresponding to the respective lane changing performance indicators, a weighted sum is performed on the respective lane changing performance indicators to obtain a total score value of the lane changing performance; the total score value of the lane changing performance is used to characterize the lane changing performance of the autonomous driving vehicle during the autonomous driving process.
7. The test system according to claim 1 or 2, characterized in that: The adaptive scenario task generator includes a scenario deconstruction module, a driving task generation module and a driving task allocation module; The scene deconstruction module is used to deconstruct the virtual traffic scene to obtain scene information for lane change testing; The driving task generating module is used to generate a plurality of different driving tasks according to the scenario information of the lane change test; The driving task allocation module is used to allocate the multiple different driving tasks to corresponding driving simulators.
8. The test system according to claim 1 or 2, characterized in that: The test evaluation unit is also used to: evaluate the performance of the manually driven vehicle during the test according to the vehicle motion information, vehicle control information, surrounding vehicle information and traffic scene information of each driving simulator during the test; Among them, the evaluation indicators for evaluating the performance of manually driven vehicles during the test process include one or more of the task completion rate, task completion time, average speed during driving, lane keeping stability, smoothness of steering wheel control, number of collisions when performing tasks, and emergency braking reaction time.
9. The test system according to claim 8, characterized in that: The evaluation index of lane keeping stability includes one or more of lane departure frequency and amplitude, lane position consistency and steering control accuracy.
Citation Information
Patent Citations
Test method and device of automatic driving system, electronic equipment and storage medium
CN115326422A
Automatic driving test system based on digital twinborn virtual-real combination and test method thereof
CN116755954A
Method and system for testing autonomous vehicle
CN118376420A
Method and system for the virtual verification of automated driving functions
DE102022132922A1
Automatic driving testing method and system, vehicle and storage medium
WO2022183449A1
Cited By
Method and device for testing intelligent networked automobile in electromagnetic compatibility darkroom and medium
CN120214473A
Automatic driving NOA third-party test and evaluation system
CN120356180A
