Performance test method and system for unmanned vehicle

Abstract

The embodiment of the invention provides a performance test method and system of an unmanned vehicle, and relates to the technical field of unmanned driving, a performance test program is preset in a vehicle control system of the unmanned vehicle, and a test process is defined in the performance test program. The testing process comprises the steps that the unmanned vehicle is controlled to be accelerated at the preset accelerator opening degree from the static state, and when the vehicle speed reaches the brake triggering speed, the unmanned vehicle is controlled to stop accelerating and conduct autonomous braking at the target deceleration till the unmanned vehicle is static. When the performance test is carried out, the values of the control parameters in the performance test program are calibrated, and the test starting instruction is sent to the unmanned vehicle, so that the vehicle control system can autonomously execute the test flow defined in the performance test program, and the performance test of the unmanned vehicle is realized. According to the embodiment of the invention, the method can achieve the efficient performance testing of the unmanned vehicle.

Description

Technical Field

[0001] This application relates to the field of autonomous driving technology, and in particular to performance testing methods and systems for driverless vehicles. Background Technology

[0002] When testing core performance characteristics of traditional driver-cab vehicles, such as acceleration and braking, a safety officer sitting in the cab typically leads the testing process and ensures vehicle safety. In actual testing, however, driving commands are usually issued manually via a host computer to execute the tests, allowing for real-time intervention and troubleshooting based on any abnormalities during vehicle operation.

[0003] However, with the widespread adoption of driverless vehicles, the testing model that allows for human intervention and control has become obsolete, leading to a fundamental shift in performance testing. In response to this shift, there is an urgent need for a new performance testing paradigm for driverless vehicles. Summary of the Invention

[0004] The purpose of this application is to provide a performance testing method and system for autonomous vehicles, so as to achieve efficient performance testing of autonomous vehicles. The specific technical solution is as follows: In a first aspect, embodiments of this application provide a performance testing method for an autonomous vehicle, including: The values ​​of control parameters within the performance test program are calibrated. The performance test program is pre-set in the vehicle control system of the autonomous vehicle under test, and the control parameters include braking trigger speed and target deceleration. A test start command is issued to the autonomous vehicle via remote control, so that the vehicle control system responds to the test start command and executes the test procedure defined in the performance test program according to the calibrated control parameters. The test procedure includes: controlling the autonomous vehicle to accelerate from a stationary state with a preset throttle opening, and when the autonomous vehicle accelerates to the braking trigger speed, controlling the autonomous vehicle to stop accelerating and autonomously brake at the target deceleration until the autonomous vehicle comes to a stop; acquiring the driving data recorded by the autonomous vehicle during the execution of the test procedure; and analyzing the acceleration and braking performance of the autonomous vehicle based on the driving data.

[0005] Secondly, embodiments of this application provide a performance testing system for an autonomous vehicle, including an autonomous vehicle, a remote controller, and a test control and analysis device. The test control and analysis device is used to calibrate the values ​​of control parameters within the performance testing program. The performance testing program is preset within the vehicle control system of the autonomous vehicle, and the control parameters include braking trigger speed and target deceleration. The remote controller is used to issue a test start command to the autonomous vehicle via remote control. The vehicle control system of the autonomous vehicle is used to respond to the test start command and execute the test process defined within the performance testing program according to the calibrated control parameters. The test process includes: controlling the autonomous vehicle to accelerate from a stationary state with a preset throttle opening, and when the autonomous vehicle accelerates to a speed reaching the braking trigger speed, controlling the autonomous vehicle to stop accelerating and autonomously brake with a target deceleration until the autonomous vehicle comes to a stop. The test control and analysis device is also used to acquire driving data recorded by the autonomous vehicle during the execution of the test process; and to analyze the acceleration and braking performance of the autonomous vehicle based on the driving data.

[0006] Beneficial effects of the embodiments in this application: The performance testing method and system for autonomous vehicles provided in this application include a pre-set performance testing program within the vehicle control system of the autonomous vehicle, and a test procedure specifically designed for autonomous vehicles defined within the performance testing program. The test procedure includes: controlling the autonomous vehicle to accelerate from a stationary state with a preset throttle opening to bring the autonomous vehicle into the acceleration phase; when the autonomous vehicle accelerates to a speed that reaches the braking trigger speed, controlling the autonomous vehicle to stop accelerating and autonomously braking with a target deceleration to transition the autonomous vehicle from the acceleration phase to the braking phase, until the autonomous vehicle comes to a stop.

[0007] Because the complete execution logic of the testing process is embedded within the vehicle control system as a performance test program, during actual performance testing of autonomous vehicles, it is only necessary to calibrate the values ​​of the control parameters within the performance test program and issue a test start command to the autonomous vehicle to instruct the vehicle control system to initiate the test process. The vehicle control system can then autonomously execute a series of operations defined in the complete test process based on this performance test program, achieving fully automated performance testing of autonomous vehicles. It can be seen that this performance testing process is simple to operate, easy to repeat, and highly efficient. In scenarios requiring performance testing of a large number of autonomous vehicles, it can significantly reduce the time required for batch testing.

[0008] Furthermore, since the test process designed in this application includes a continuous acceleration phase and a braking phase, only a single test process is needed to test both the acceleration and braking performance of the autonomous vehicle, without having to test the acceleration and braking performance of the autonomous vehicle separately through different operation processes. This can improve the efficiency of testing the acceleration and braking performance of the autonomous vehicle.

[0009] Of course, implementing any product or method of this application does not necessarily require achieving all of the advantages described above at the same time. Attached Figure Description

[0010] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other embodiments can be obtained based on these drawings.

[0011] Figure 1 A flowchart illustrating the performance testing method for an unmanned vehicle provided in an embodiment of this application; Figure 2 A schematic diagram illustrating a performance testing process provided in an embodiment of this application; Figure 3 This is a schematic diagram of the execution logic when the VCU performs the test process according to the embodiments of this application. Detailed Implementation

[0012] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art based on this application are within the scope of protection of this application.

[0013] When testing core performance characteristics of traditional driver-cab vehicles, such as acceleration and braking, a safety officer sitting in the cab typically leads the testing process and ensures vehicle safety. In actual testing, however, driving commands are usually issued manually via a host computer to execute the tests, allowing for real-time intervention and troubleshooting based on any abnormalities during vehicle operation.

[0014] However, with the widespread adoption of driverless vehicles, the testing model that allows for human intervention and control has become obsolete, leading to a fundamental shift in performance testing. In response to this shift, there is an urgent need for a new performance testing paradigm for driverless vehicles.

[0015] Taking mining transportation as an example, wide-body dump trucks are core equipment for efficient transportation in both open-pit and underground mines. With the development of smart mines and unmanned operations, traditional wide-body dump trucks with cabs are rapidly evolving into driverless vehicles without cabs. While this design, which removes the cab and completely eliminates the onboard safety operator, is revolutionary in terms of improving inherent safety, reducing labor costs, and optimizing vehicle structure, it also brings new and severe challenges to the performance testing of wide-body dump trucks for the reasons mentioned above.

[0016] In view of this, embodiments of this application provide a performance testing method for autonomous vehicles, used to conduct performance tests on autonomous vehicles. In one example, the autonomous vehicle for which this method is applied to the performance testing can be a cab-less wide-body mining dump truck. See also... Figure 1 The performance testing method for this autonomous vehicle specifically includes the following steps: Step S101: Calibrate the values ​​of the control parameters in the performance test program; the performance test program is preset in the vehicle control system of the unmanned vehicle to be tested, and the control parameters include braking trigger speed and target deceleration.

[0017] Specifically, this application embodiment designs a dedicated testing procedure for autonomous vehicles (the details of which will be described in more detail later). By following this testing procedure to control the operation of the autonomous vehicle, the acceleration and braking performance of the autonomous vehicle can be tested. To ensure the efficiency and repeatability of the testing process, this application embodiment incorporates this testing procedure as a performance test program into the vehicle control system of the autonomous vehicle under test. Thus, the vehicle control system can automatically execute the testing procedure based on the performance test program.

[0018] The performance testing program defines the complete operational logic of the test process, and this operational logic includes control parameters with adjustable values. After the values ​​of the control parameters are set, the operational logic defined in the performance testing program can be executed on the actual vehicle. In this embodiment, the control parameters in the performance testing program include: braking trigger speed and target deceleration.

[0019] The values ​​of control parameters within the performance testing program are calibrated, which involves writing the set values ​​of the control parameters into the performance testing program. In practical applications, the values ​​of the control parameters can be set according to specific performance testing requirements, and this embodiment does not impose any restrictions on this.

[0020] In one example, the vehicle control system in this application embodiment can be understood as the VCU (Vehicle Control Unit) of an autonomous vehicle.

[0021] Step S102: Send a test start command to the autonomous vehicle via remote control so that the vehicle control system responds to the test start command and executes the test process defined in the performance test program according to the calibrated control parameters. The test process includes: controlling the autonomous vehicle to accelerate from a stationary state with a preset throttle opening, and when the autonomous vehicle accelerates to the speed that reaches the braking trigger speed, controlling the autonomous vehicle to stop accelerating and autonomously brake with a target deceleration until the autonomous vehicle comes to a stop.

[0022] After calibrating the control parameters in the performance test program based on the aforementioned step S101, a test start command can be issued to the autonomous vehicle under test via remote control to initiate the performance test for that vehicle. It should be noted that before issuing the test start command to the autonomous vehicle, it must be ensured that the vehicle is parked in a pre-selected test area. This application embodiment does not limit the selection method of the test area; it can be chosen according to actual needs. In one example, the path width within the test area can be no less than twice the size of the autonomous vehicle, and the effective straight section length can be greater than 50 meters to meet the basic dynamic space requirements for vehicle acceleration and braking.

[0023] In one possible implementation of this application, the test start command can be issued by the tester via a remote control. This application does not limit the specific form of the remote control; it can be selected according to actual needs. For example, the remote control can be a handheld emergency stop remote control device with a line-of-sight communication range of 200 meters. In one example, the remote control has a test start button, which the tester can press to issue the test start command. Furthermore, to prevent the tester from accidentally issuing the test start command by pressing the test start button, an activation confirmation duration (e.g., 5 seconds) can be configured for triggering the test start command. Therefore, the tester needs to press and hold the test start button for more than the activation confirmation duration to issue the test start command.

[0024] After receiving a test start command, the vehicle control system of an autonomous vehicle can respond to the test start command and execute the test process defined in the performance test program according to the calibrated control parameters to achieve automated testing of the autonomous vehicle.

[0025] Based on the description of the test procedure in step S102, it can be seen that the test procedure designed in this application embodiment actually includes two consecutive stages: the first is the autonomous linear acceleration stage (acceleration stage), and the second is the autonomous braking stage (braking stage). Specifically, the autonomous vehicle starts from a stationary state and accelerates at a preset throttle opening, thus entering the acceleration stage; when it is determined that the autonomous vehicle has accelerated to a speed that reaches the braking trigger speed, and the autonomous vehicle is controlled to stop accelerating and autonomously brake at a target deceleration, the autonomous vehicle ends the acceleration stage and enters the braking stage, which stops when the autonomous vehicle brakes to a standstill.

[0026] It is understood that, since the test process designed in this application embodiment includes a sequential acceleration phase and a braking phase, both the acceleration performance and braking performance of the autonomous vehicle can be tested based on this single test process. Specifically, the acceleration phase is used to test the acceleration performance of the autonomous vehicle, and the braking phase is used to test the braking performance of the autonomous vehicle.

[0027] Therefore, in the aforementioned step S101, the values ​​of the braking trigger speed and target deceleration in the control parameters can be set according to the actual test requirements for acceleration and braking performance. For example, the braking trigger speed can be set to 30 km / h. Furthermore, the preset throttle opening value in the test procedure can also be set according to actual needs, for example, it can be set to 80%. Of course, in practical applications, the throttle opening can also be included as an adjustable control parameter. Testers can set the throttle opening value according to the actual acceleration performance test requirements, and in step S101, the throttle opening value in the performance test program will be calibrated to the set value.

[0028] As mentioned earlier, the vehicle control system in this embodiment can be understood as the VCU of an autonomous vehicle. From the perspective of VCU execution, this embodiment edits the designed test process into the VCU in the form of a performance test program, which is equivalent to solidifying the designed test process into a digital instruction sequence that can be executed by the VCU. Specifically, based on the edited performance test program, the VCU can respond to the test start command and autonomously send an acceleration command to the acceleration system to accelerate at a preset throttle opening, so that the autonomous vehicle enters the acceleration phase; and when it is determined that the autonomous vehicle has accelerated to the braking trigger speed, it autonomously resets the acceleration command to zero and sends a braking command to the braking system to brake at the target deceleration, so that the autonomous vehicle transitions from the acceleration phase to the braking phase.

[0029] Step S103: Obtain the driving data recorded by the autonomous vehicle during the execution of the test process.

[0030] In one example, a high-precision data logger can be deployed on an autonomous vehicle to synchronously collect bus and sensor data throughout the entire testing process. After the testing process concludes, the data recorded by the high-precision logger is retrieved as driving data.

[0031] Step S104: Analyze the acceleration and braking performance of the autonomous vehicle based on the driving data.

[0032] Specifically, acceleration performance indicators can be determined based on driving data during the acceleration phase of the autonomous vehicle testing process, enabling quantitative analysis of the vehicle's acceleration performance. The specific indicators can be selected according to actual performance testing needs; for example, acceleration performance indicators may include one or more of acceleration distance and acceleration time.

[0033] Similarly, braking performance indicators for autonomous vehicles can be determined based on driving data during the braking phase of the testing process, enabling quantitative analysis of their braking performance. Specific indicators can be selected based on actual performance testing needs. For example, braking performance indicators may include one or more of braking distance, braking duration, and braking control accuracy. Braking control accuracy may include braking speed control accuracy, which reflects the deviation between the actual deceleration and the target deceleration during the braking phase.

[0034] As mentioned earlier, configuring performance testing programs within the VCU is equivalent to solidifying the designed test process into a sequence of digital instructions that can be executed by the VCU. In one possible implementation of this application, based on the analysis of the acceleration and braking performance of the autonomous vehicle using driving data, the system response characteristics of the autonomous vehicle can be further analyzed using driving data. For example, specifically, the response delay of the acceleration system and the response delay of the braking system of the autonomous vehicle can be determined based on the driving data.

[0035] The performance testing method for autonomous vehicles provided in this application includes a pre-set performance testing program within the vehicle control system of the autonomous vehicle, and defines a test procedure specifically designed for autonomous vehicles within the performance testing program. The test procedure includes: controlling the autonomous vehicle to accelerate from a stationary state with a preset throttle opening to bring the autonomous vehicle into the acceleration phase; when it is determined that the autonomous vehicle has accelerated to a speed that reaches the braking trigger speed, controlling the autonomous vehicle to stop accelerating and autonomously braking with a target deceleration to transition the autonomous vehicle from the acceleration phase to the braking phase, until the autonomous vehicle comes to a stop.

[0036] Because the complete execution logic of the testing process is embedded within the vehicle control system as a performance test program, during actual performance testing of autonomous vehicles, it is only necessary to calibrate the values ​​of the control parameters within the performance test program and issue a test start command to the autonomous vehicle to instruct the vehicle control system to initiate the test process. The vehicle control system can then autonomously execute a series of operations defined in the complete test process based on this performance test program, achieving fully automated performance testing of autonomous vehicles. It can be seen that this performance testing process is simple to operate, easy to repeat, and highly efficient. In scenarios requiring performance testing of a large number of autonomous vehicles, it can significantly reduce the time required for batch testing.

[0037] Furthermore, since the test process designed in this application includes a continuous acceleration phase and a braking phase, only a single test process is needed to test both the acceleration and braking performance of the autonomous vehicle, without having to test the acceleration and braking performance of the autonomous vehicle separately through different operation processes. This can improve the efficiency of testing the acceleration and braking performance of the autonomous vehicle.

[0038] In practical applications, the aforementioned performance testing methods for unmanned vehicles can be used to test the performance of cab-less wide-body mining dump trucks to ensure that the core motion control system has high reliability and stability before the delivery of the wide-body mining dump truck, thereby ensuring the safety and efficiency of unmanned transportation operations in mines.

[0039] In one embodiment of this application, before performing the aforementioned step S102, the driverless vehicle to be tested can be controlled to enter an autonomous driving ready state, and the driverless vehicle can be ensured to be in the initial test state. The driverless vehicle being in the initial test state means that the driverless vehicle is in a parked state and the transmission is in neutral.

[0040] Specifically, before issuing a test start command to the autonomous vehicle, by controlling the vehicle to enter an autonomous driving ready state and ensuring it is in its initial state, it is guaranteed that the autonomous vehicle remains stationary before the test process begins and can respond instantly to commands from the vehicle control system. This ensures that the autonomous vehicle can respond promptly to the test start command and begin operation, and guarantees the effectiveness of the acceleration phase during the test process.

[0041] Correspondingly, before controlling the autonomous vehicle to accelerate from a standstill at a preset throttle opening, the test procedure defined within the performance test program also includes: controlling the autonomous vehicle to release the parking brake and switching the transmission to a forward gear. Specifically, in the test procedure, after confirming that the autonomous vehicle's parking brake has been released and the transmission has been engaged in a forward gear, the operation of controlling the autonomous vehicle to accelerate from a standstill at a preset throttle opening is then performed.

[0042] From the perspective of VCU execution, VCU can respond to the test start command, first autonomously execute the operation sequence of releasing the handbrake and engaging the forward gear. After confirming that the parking brake has been released and the transmission has been engaged in the forward gear, it then autonomously sends an acceleration command to the acceleration system to accelerate at a preset throttle opening, so that the autonomous vehicle enters the acceleration phase.

[0043] In one embodiment of this application, after controlling the autonomous vehicle to stop accelerating and autonomously brake at a target deceleration, the test process defined in the performance test program further includes: after determining that the speed of the autonomous vehicle has dropped to no more than a preset parking speed and maintaining a preset observation time, controlling the autonomous vehicle to enter the test initial state.

[0044] In other words, during the braking phase of the test process, once it is determined that the speed of the autonomous vehicle has dropped to no more than the preset parking speed, and the preset observation time is maintained while the speed is not below the preset parking speed, the parking brake can be reapplied to put the autonomous vehicle into a parking state, and the transmission of the autonomous vehicle can be put into neutral, so that the autonomous vehicle can return to the initial test state.

[0045] The preset parking speed and preset observation time can be set according to actual needs, and this application embodiment does not limit them. In one example, the preset parking speed can be set to 0.1 km / h and the preset observation time to 2 seconds.

[0046] As can be easily understood from the preceding explanation, the test procedures defined in the performance test program need to be performed under the condition that the autonomous vehicle is in the initial test state. When defining the test procedures within the performance test program according to the description in the embodiments of this application, the vehicle control system can automatically restore the autonomous vehicle to the initial test state after controlling the autonomous vehicle to enter the acceleration and braking phases sequentially, thereby eliminating the need for additional operations to restore it to the initial test state before conducting the next performance test on the autonomous vehicle, which helps to improve the testing efficiency in iterative testing scenarios.

[0047] In one embodiment of this application, after controlling the autonomous vehicle to accelerate from a stationary state with a preset throttle opening, the test process further includes: if the speed of the autonomous vehicle has not reached the braking trigger speed after accelerating for a preset waiting time, then when it is determined that the autonomous vehicle has accelerated for the preset waiting time, the operation of controlling the autonomous vehicle to stop accelerating and perform autonomous braking with a target deceleration is executed.

[0048] Specifically, in actual testing scenarios, factors such as insufficient vehicle power due to an excessively high braking trigger speed may prevent the autonomous vehicle from accelerating to the braking trigger speed. The above-mentioned design of the testing process in this application embodiment is to avoid this phenomenon, which would prevent the autonomous vehicle from entering the braking phase and thus prevent the braking performance of the autonomous vehicle from being tested.

[0049] Specifically, in this embodiment of the application, if the vehicle speed of the autonomous vehicle has not reached the braking trigger speed after entering the preset waiting time of the acceleration phase during the execution of the test process according to the performance test procedure, the vehicle control system will execute the operation of "controlling the autonomous vehicle to stop accelerating and autonomously braking with the target deceleration" in the test process as soon as it is determined that the autonomous vehicle has entered the preset waiting time of the acceleration phase, instead of waiting until the vehicle speed reaches the braking trigger speed before executing the operation.

[0050] The preset waiting time can be set according to actual needs, and this application embodiment does not impose any restrictions on it. In one example, the preset waiting time can be set to 10 seconds.

[0051] By defining the test flow within the performance testing procedure according to the description of the embodiments of this application, it is possible to control the autonomous vehicle to directly enter the braking phase if the vehicle speed has entered the acceleration phase and has not yet reached the braking trigger speed after a preset waiting time. This ensures that even if the autonomous vehicle's speed cannot accelerate to the braking trigger speed, the autonomous vehicle can still smoothly enter the braking phase, guaranteeing that the braking performance test of the autonomous vehicle is not affected.

[0052] Specifically, each time the vehicle control system executes the test procedures defined in the performance test program for an autonomous vehicle, it constitutes a single test cycle for that vehicle. Therefore, by controlling the vehicle control system to execute the test procedures defined in the performance test program multiple times for the autonomous vehicle, iterative testing of the autonomous vehicle can be achieved.

[0053] In one embodiment of this application, in order to implement iterative testing of autonomous vehicles, after issuing a test start command to the autonomous vehicle based on the aforementioned step S102 to instruct the vehicle control system to start the test process, a test stop command can be issued to the autonomous vehicle via remote control after the corresponding test process ends (as can be easily understood from the above description, the test process specifically ends when the autonomous vehicle brakes to a standstill), so as to reset the performance test program configured in the vehicle control system to the initial state.

[0054] The term "reset" here can be understood as a software reset. Specifically, by resetting the performance test program to its initial state, program anomalies can be avoided during the next run of the performance test program due to residual states from the previous run.

[0055] In this embodiment of the application, in order to facilitate the next test cycle of the driverless vehicle, in addition to restoring the performance test program in the vehicle control system to the initial state, it is also necessary to control the driverless vehicle to turn around or return to its original position through remote control.

[0056] Specifically, controlling the autonomous vehicle to turn around involves using the endpoint of the vehicle's journey in one test cycle as the starting point for the next test cycle, allowing the vehicle to travel back and forth on the road for iterative testing. Controlling the autonomous vehicle to return to its original position means guiding it back to the starting point of the previous test cycle, allowing it to start from the same location on the road and travel in the same direction for iterative testing. In practical applications, the choice between these two approaches can be made based on specific requirements.

[0057] After the performance test program has been reset to its initial state and the autonomous vehicle has turned around or returned to its original position, the next test cycle for the autonomous vehicle can be initiated by performing either the first or second operation as follows: First operation: Send a test start command to the driverless vehicle again via remote control, so that the vehicle control system responds to the resent test start command and executes the test process defined in the performance test program according to the calibrated control parameters.

[0058] In the first operation, specifically after the performance test program is reset to its initial state and the autonomous vehicle has turned around or returned to its original position, a test start command is directly issued to the autonomous vehicle. In this case, the values ​​of the control parameters within the performance test program are the values ​​calibrated last time. Therefore, when executing the first operation to implement the next test cycle for the autonomous vehicle, the vehicle control system uses the same values ​​of performance parameters in both the preceding and following test processes.

[0059] The second step is to recalibrate the values ​​of the control parameters within the performance test program; and to remotely send a test start command to the unmanned vehicle so that the vehicle control system responds to the re-sent test start command and executes the test process defined in the performance test program according to the recalibrated control parameters.

[0060] In the second operation, the control parameters can be set to values ​​different from the previous values ​​according to actual needs. Then, the control parameters in the performance test program are recalibrated based on the newly set values. After recalibration, the test start command is issued to the autonomous vehicle. Therefore, when the second operation is performed to implement the next test cycle for the autonomous vehicle, the performance parameter values ​​used by the vehicle control system in the two test processes are different.

[0061] In practical applications, if an iterative testing scheme involving N test cycles is required for autonomous vehicles, the iterative scheme can be achieved by starting the first test cycle according to the aforementioned steps S101-S102, and executing the aforementioned first or second operation at the end of each test cycle in the first N-1 test cycles. In one possible implementation of this application, data reading and performance analysis can be performed after all test processes required for all test cycles in the iterative testing scheme have been completed, based on the aforementioned steps S103-S104, to save on the processing costs of data reading and performance analysis.

[0062] It is understandable that real-world vehicle performance testing scenarios place extremely high demands on test safety. Without immediate human intervention, any malfunction in a test case could lead to loss of vehicle control, resulting in significant safety risks and economic losses. Therefore, performance testing must be conducted in a highly controlled environment.

[0063] Considering the aforementioned requirements for test safety, in one embodiment of this application, an emergency stop control permission can be configured for the remote controller used to issue the test start command, providing the highest level of safety redundancy. Specifically, relevant personnel can be arranged to observe the actual execution of the test process at the test site, and when a safety risk is noticed to the autonomous vehicle, an emergency stop command can be issued to the autonomous vehicle remotely via the remote controller. This causes the vehicle control system of the autonomous vehicle to respond to the emergency stop command and perform a preset emergency braking operation on the autonomous vehicle.

[0064] The preset emergency braking operation is designed to bring the autonomous vehicle to a stop as quickly as possible within the shortest possible distance. The specific details of this preset emergency braking operation can be set according to actual conditions. For example, the preset emergency braking operation could be controlling the autonomous vehicle to brake with a preset emergency braking force. The magnitude of the emergency braking force can be set according to actual needs; for instance, the emergency braking force can be set to the maximum braking force achievable by the braking system, or 90% of that maximum braking force.

[0065] In one embodiment of this application, in order to prevent the execution of the test process from being interfered with by the relevant personnel due to the misoperation of the remote control, the remote control can be locked during the execution of the test process, except for the emergency stop control authority, so as to fundamentally prevent interference from misoperation.

[0066] In one embodiment of this application, to ensure the absolute safety of the performance testing process and the validity of the results, the above-described performance testing method for autonomous vehicles may enforce the following core specifications during the execution of the testing process: Core Specification 1: When the autonomous vehicle being tested is a new energy vehicle, it must be ensured that the high-voltage system of the entire vehicle is powered on; when the autonomous vehicle being tested is a diesel-powered vehicle, it must be ensured that the engine speed is higher than the preset minimum operating speed.

[0067] The first core specification is the prerequisite for the power system to execute the test process. Among them, the preset minimum execution speed can be set according to actual needs. For example, the minimum execution speed can be set to 600 RPM (Revolutions Per Minute).

[0068] Core Specification 2: Ensure that the fault level of the entire autonomous vehicle is no higher than Level 2.

[0069] The second core specification is the health status prerequisite for executing the testing process. Specifically, the determination of the overall vehicle fault level for autonomous vehicles can refer to standards such as the "Regulations on Technical Management of Vehicles in the Road Transport Industry" or "Failure Modes and Classifications of Automobiles," which will not be elaborated upon in this embodiment.

[0070] In one possible implementation of this application, to ensure that the aforementioned core specifications are enforced, emergency stop conditions can be defined within the performance testing program of the vehicle control system. The emergency stop conditions include at least one of the following first, second, and third conditions: First condition: When the driverless vehicle is a new energy vehicle, the first condition is that the high-voltage system of the driverless vehicle is not powered on; when the driverless vehicle is a diesel-powered vehicle, the first condition is that the engine speed of the driverless vehicle is lower than the preset minimum execution speed.

[0071] The second condition is that the overall fault level of the driverless vehicle is higher than level two.

[0072] The third condition is that the vehicle control system receives an emergency stop command issued via remote control.

[0073] Based on the aforementioned emergency stop conditions defined within the performance test program, the following emergency stop logic is specifically defined within the performance test program: when any one of the emergency stop conditions is detected to be met during the execution of the test process, a preset emergency braking operation is performed on the autonomous vehicle.

[0074] The specific details of the preset emergency execution operations can be found in the explanation above.

[0075] Thus, when the vehicle control system detects that any one of the emergency stop conditions is met during the execution of the test process defined in the performance test program, it can autonomously perform a preset emergency braking operation on the unmanned vehicle based on the emergency stop logic defined in the performance test program.

[0076] As can be seen, in the embodiments of this application, the first and second conditions in the emergency conditions are used to ensure that the aforementioned core specification one and core specification two are enforced, while the third condition is used to implement the emergency stop control authority of the remote control.

[0077] In one embodiment of this application, during the process of controlling an autonomous vehicle to brake autonomously at a target deceleration, the vehicle control system specifically sets the braking force of the autonomous vehicle based on a preset reference calibration table to control the deceleration of the autonomous vehicle. The reference calibration table provides a mapping relationship between the braking force and the deceleration achievable under the braking force. This application does not limit the source of the reference calibration table; in one example, the reference calibration table may be set empirically, and in another example, it may be based on the braking performance test results of other similar vehicle models.

[0078] In this embodiment of the application, when the braking performance determined in the aforementioned step S104 includes braking deceleration control accuracy, the mapping relationship given in the reference calibration table can also be updated based on the determined braking deceleration control accuracy.

[0079] Specifically, if the braking deceleration control accuracy indicates that the actual deceleration is smaller than the target deceleration, it means that the braking force determined based on the benchmark calibration table is larger than the actual braking force required to achieve the target deceleration. In this case, the braking force within the mapping relationship given in the benchmark calibration table can be reduced. Similarly, if the braking deceleration control progress table indicates that the actual deceleration is larger than the target deceleration, it means that the braking force determined based on the benchmark calibration table is smaller than the actual braking force required to achieve the target deceleration. In this case, the braking force within the mapping relationship given in the benchmark calibration table can be increased. This application does not limit the specific method of adjusting the braking force in the benchmark calibration table. In one example, the conversion relationship between deceleration deviation and braking force deviation can be preset. First, the deceleration deviation is determined based on the braking deceleration control accuracy. Then, the determined deceleration deviation is converted into a braking force deviation according to the conversion relationship. The magnitude of the braking force deviation is then adjusted based on the determined braking force deviation.

[0080] In this embodiment, by updating the mapping relationship given in the basic calibration table based on the braking deceleration control accuracy obtained from the test, the autonomous vehicle can achieve higher braking deceleration control accuracy when braking based on the basic calibration table during subsequent driving, which is beneficial to improving driving safety. After updating the mapping relationship given in the reference calibration table, the autonomous vehicle can be re-performed based on the aforementioned steps S101-S104 to confirm whether its braking speed control accuracy has been improved. If the deviation and adjustment range are not large, the autonomous vehicle can be put into subsequent use.

[0081] To facilitate understanding of the above embodiments of this application, the following is an exemplary description of a possible performance testing process when using the performance testing method for unmanned vehicles provided in this application to perform performance testing on a wide-body dump truck used in mining: The performance testing process is as follows: Figure 2As shown. During performance testing, it should first be confirmed that the environmental conditions for the performance test meet the requirements, that is, ensuring that the path width is not less than twice the vehicle width and the effective straight section length is greater than 50 meters. After confirming that the environmental conditions meet the requirements and the vehicle posture of the mining wide-body dump truck is correct, the control parameters in the pre-set performance test program in the VCU are calibrated using calibration software, with the CAN (Controller Area Network) box as the data transmission medium. Specifically, the automatic trigger speed and target deceleration need to be calibrated. After completing the calibration of the control parameters, a test start command is issued via a handheld remote control, allowing the VCU to autonomously execute the test process defined in the performance test program in response to the test start command. After the test process is completed, a test stop command is issued via the handheld remote control to restore the performance test program to its initial state and control the mining wide-body dump truck to turn around or return to its original position for the next round of testing. After the test is completed, data can be acquired through a data recorder, and the acceleration and braking performance of the mining wide-body dump truck can be analyzed based on the acquired data. Furthermore, throughout the entire testing process executed by the VCU, an emergency stop command can be issued via a handheld remote control to enable the VCU to control the mining wide-body dump truck to stop suddenly, thus ensuring the safety of the test.

[0082] The execution logic of VCU when autonomously executing the test process within the performance test program is as follows: Figure 3 As shown. Specifically, the mining wide-body dump truck is in the initial test state before the test process begins, and the following conditions are met ( Figure 3 (The '&' symbol indicates an AND relationship) indicates that the wide-body dump truck for mining has entered the ready state. Remote start flag == 1; Vehicle speed == 0; Actual parking status == Parking; Current gear == Neutral; High voltage status == 1; Vehicle fault level <= 2.

[0083] Specifically, when the VCU receives a test start command, the remote start flag is 1; when the VCU does not receive a test stop command, the remote start flag is 0. "Actual Parking Status == Parking" indicates that the mining wide-body dump truck is in a parked state. "Current Gear" refers to the gearbox gear. "High Voltage Status == 1" indicates that the entire mining wide-body dump truck is connected to high voltage; if "High Voltage Status == 0", it indicates that the entire mining wide-body dump truck is not connected to high voltage.

[0084] When any of the following conditions are met ( Figure 3 (The symbol "||" in the text indicates an OR relationship). The mining wide-body dump truck returns from the ready state to the test initial state: Remote start flag == 0; Remote stop flag == 1; High voltage status = 0; Vehicle fault level > 2.

[0085] When the VCU receives an emergency stop command, the remote stop flag is set to 1.

[0086] When the mining wide-body dump truck is in a ready state, if all the conditions in "Remote start flag == 1; Vehicle speed == 0; Actual parking status == Parking; Current gear == Neutral; High voltage status == 1; Vehicle fault level <= 2" are met for 5 seconds, the VCU will perform the operation of engaging a forward gear and releasing the handbrake (releasing the handbrake is equivalent to releasing the parking brake): After the VCU performs the operations of engaging a drive gear and releasing the handbrake, the VCU controls the acceleration system to accelerate at 80% throttle opening when all of the following conditions are met: Actual parking status == Parking released; Current gear == Forward gear; High voltage status == 1; Vehicle fault level <= 2.

[0087] Among them, "Actual parking status == parking brake released" means that the parking brake has been released.

[0088] After the VCU-controlled acceleration system accelerates at 80% throttle opening, when all conditions in "vehicle speed >= brake trigger speed; high voltage status == 1; vehicle fault level <= 2" are met, the VCU will autonomously reset the throttle to 0 (i.e., stop acceleration) and send the target deceleration to the braking system. Furthermore, if the mining wide-body dump truck has maintained acceleration at 80% throttle opening for 10 seconds, the VCU will also autonomously reset the throttle to 0 and send the target deceleration to the braking system after maintaining this state for 10 seconds.

[0089] After the VCU clears the throttle to 0 and issues the target deceleration to the braking system, if all the conditions in "vehicle speed <= 0.1 km / h; high pressure status == 1; vehicle fault level <= 2" are met for 2 seconds, the VCU will clear the deceleration to 0, put the transmission in neutral, and engage the handbrake.

[0090] After the VCU clears the deceleration to 0, puts the transmission in neutral, and engages the handbrake, the mining wide-body dump truck returns to the initial test state when either the "remote start flag == 0" or the "remote stop flag == 1" condition is met.

[0091] Furthermore, from the moment the VCU engages the forward gear and releases the handbrake until the VCU decelerates to zero, and the transmission is shifted to neutral and the handbrake is engaged, the VCU will immediately execute protective measures (that is, perform a preset emergency braking operation for the mining wide-body dump truck) if any of the following conditions are met: Remote stop flag == 1; High voltage status == 0; Vehicle fault level > 2.

[0092] In summary, the performance testing method for autonomous vehicles provided in this application has the following advantages: 1. A fundamental innovation in testing paradigm: This performance testing method completely eliminates the reliance on subjective driver operation and evaluation in traditional manned driving tests, establishing a fully automated closed-loop testing paradigm characterized by "programmable instructions, monitorable processes, and quantifiable results." By solidifying the testing logic into a precisely repeatable sequence of digital instructions and replacing manual perception with high-precision data recording, a revolutionary shift from experience-driven to data-accelerated testing is achieved, ensuring the objectivity, consistency, and traceability of the test.

[0093] 2. A multi-redundant safety assurance system has been constructed: This performance testing method does not simply remove human intervention, but rather designs a systematic proactive safety boundary. Through strict pre-processing state verification (such as road conditions, vehicle mode, and fault level), intelligent isolation of the remote control's regular control permissions during execution (to prevent accidental touches), and the retention of the highest priority hardware emergency stop redundancy, a multi-layered protection mechanism combining "software logic limits" and "hardware emergency intervention" has been constructed. This significantly enhances the inherent safety of the testing process while improving the level of automation.

[0094] 3. Establishment of standardized and repeatable test benchmarks: This performance testing method establishes a rigorous standardized benchmark for performance testing through precise environmental requirements, unified initial vehicle state settings, preset calibration parameters, and fixed trigger speeds and command thresholds. This ensures that each test is conducted under identical conditions, eliminating errors introduced by human variables. The generated performance data (such as response latency and deceleration accuracy) is highly comparable, providing a reliable basis for accurate benchmarking, consistent evaluation, and continuous optimization of vehicle performance.

[0095] 4. Significantly improves testing efficiency and engineering practicality: The automated process realizes a standardized operation across the entire chain from preparation and execution to reset, resulting in a short single test cycle and enabling rapid, continuous, multi-round iterative testing. This method has low dependence on existing hardware (handheld remote control, data logger) and is easy to integrate, which not only reduces implementation costs but also makes it easy to quickly deploy and apply in various scenarios such as R&D verification, factory testing, and on-site maintenance, demonstrating strong engineering practicality.

[0096] 5. Possesses broad adaptability to power systems and vehicle models: The core testing logic of this performance testing method focuses on the vehicle's drive-by-wire response and actuator performance. By clearly distinguishing and defining the starting state prerequisites (high voltage power-on / engine speed) for new energy vehicles and traditional diesel vehicles, the same methodology can be compatible with mining cabinless unmanned vehicles with different power sources, demonstrating good versatility and scalability.

[0097] Based on the same inventive concept, this application also provides a performance testing system for unmanned vehicles, which includes: an unmanned vehicle, a remote controller, and test control and analysis equipment.

[0098] The test control and analysis equipment is used to calibrate the values ​​of control parameters within the performance test program. The performance test program is pre-set in the vehicle control system of the autonomous vehicle, and the control parameters include braking trigger speed and target deceleration.

[0099] The remote controller is used to issue test start commands to driverless vehicles via remote control.

[0100] The vehicle control system of the autonomous vehicle is used to: respond to the test start command and execute the test process defined in the performance test program according to the calibrated control parameters; the test process includes: controlling the autonomous vehicle to accelerate from a stationary state with a preset throttle opening, and when it is determined that the autonomous vehicle has accelerated to the speed that reaches the braking trigger speed, controlling the autonomous vehicle to stop accelerating and autonomously braking with a target deceleration until the autonomous vehicle comes to a stop.

[0101] The test control and analysis equipment is also used to: acquire driving data recorded by the autonomous vehicle during the execution of the test process; and analyze the acceleration and braking performance of the autonomous vehicle based on the driving data.

[0102] For more information about this performance testing system and its beneficial effects, please refer to the description in the performance testing method embodiments above, which will not be repeated here.

[0103] In one embodiment of this application, the remote controller is further used to: control the driverless vehicle to enter the autonomous driving ready state via remote control, and ensure that the driverless vehicle is in the initial test state; the initial test state is: the driverless vehicle is in a parked state and the transmission is in neutral.

[0104] Before controlling the autonomous vehicle to accelerate from a standstill at a preset throttle opening, the test procedure also includes: controlling the autonomous vehicle to release the parking brake and switching the transmission to forward gear.

[0105] In the testing process, after confirming that the parking brake has been released and the transmission has been engaged in forward gear, the operation of controlling the autonomous vehicle to accelerate from a stationary state with a preset throttle opening is then performed.

[0106] In one embodiment of this application, after controlling the autonomous vehicle to stop accelerating and autonomously brake at a target deceleration, the test process further includes: after determining that the speed of the autonomous vehicle has dropped to no more than a preset parking speed and maintaining a preset observation time, controlling the autonomous vehicle to enter the test initial state.

[0107] In one embodiment of this application, after controlling the autonomous vehicle to accelerate from a stationary state with a preset throttle opening, the test process further includes: if the speed of the autonomous vehicle has not reached the braking trigger speed after accelerating for a preset waiting time, then when it is determined that the autonomous vehicle has accelerated for the preset waiting time, the operation of controlling the autonomous vehicle to stop accelerating and perform autonomous braking with a target deceleration is executed.

[0108] In one embodiment of this application, the remote controller is also used to: issue a test stop command to the unmanned vehicle via remote control to reset the performance test program configured in the vehicle control system to its initial state; and control the unmanned vehicle to turn around or return to its original position via remote control.

[0109] The remote controller is also used to: send a test start command to the driverless vehicle again via remote control, so that the vehicle control system responds to the sent test start command and executes the test process defined in the performance test program again according to the calibrated control parameters.

[0110] or, The test control and analysis equipment is also used to recalibrate the values ​​of control parameters within the performance test program. The remote controller is also used to reissue the test start command to the autonomous vehicle via remote control, so that the vehicle control system responds to the reissued test start command and executes the test procedures defined within the performance test program according to the recalibrated control parameters.

[0111] In one embodiment of this application, the performance testing program further defines an emergency stop condition, which includes at least one of the following first condition, second condition, and third condition: First condition: When the driverless vehicle is a new energy vehicle, the first condition is that the high-voltage system of the driverless vehicle is not powered on; when the driverless vehicle is a diesel-powered vehicle, the first condition is that the engine speed of the driverless vehicle is lower than the preset minimum execution speed.

[0112] The second condition is that the overall fault level of the driverless vehicle is higher than level two.

[0113] The third condition is that the vehicle control system receives an emergency stop command issued via remote control.

[0114] The vehicle control system of the autonomous vehicle is also used to perform a preset emergency braking operation on the autonomous vehicle when any of the emergency stop conditions is detected to be met during the execution of the test procedure.

[0115] In one embodiment of this application, during the execution of the test process, all control permissions of the remote controller over the autonomous vehicle except for the emergency stop control permission are locked; the emergency stop control permission is the permission of the remote controller to issue an emergency stop command to the autonomous vehicle so that the vehicle control system can control the autonomous vehicle to perform a preset emergency braking operation.

[0116] In one embodiment of this application, the test control and analysis equipment is specifically used to: determine the acceleration performance index of the autonomous vehicle based on the driving data of the autonomous vehicle during the acceleration phase, wherein the acceleration performance index includes one or more of acceleration distance and acceleration duration; and determine the braking performance index of the autonomous vehicle based on the driving data of the autonomous vehicle during the braking phase, wherein the braking performance index includes one or more of braking distance, braking duration and braking control accuracy.

[0117] The acceleration phase consists of controlling the autonomous vehicle to accelerate from a stationary state with a preset throttle opening until the vehicle reaches the braking trigger speed. The braking phase consists of controlling the autonomous vehicle to stop accelerating and autonomously braking at a target deceleration rate until the vehicle comes to a complete stop.

[0118] In one embodiment of this application, during the process of controlling the autonomous vehicle to perform autonomous braking at a target deceleration, the vehicle control system specifically sets the braking force of the autonomous vehicle based on a preset reference calibration table to control the deceleration of the autonomous vehicle; the reference calibration table is used to provide a mapping relationship between the braking force and the deceleration that can be achieved under the braking force.

[0119] The braking performance indicators determined by the test, control, and analysis equipment include braking control accuracy, which in turn includes braking deceleration control accuracy. The test, control, and analysis equipment is also used to update the mapping relationships given in the benchmark calibration table based on the determined braking deceleration control accuracy.

[0120] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0121] The various embodiments in this specification are described in a related manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the embodiment of the performance testing system for autonomous vehicles is basically similar to the method embodiment, so the description is relatively simple; relevant parts can be referred to the description of the method embodiment.

[0122] The above description is merely a preferred embodiment of this application and is not intended to limit the scope of protection of this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application are included within the scope of protection of this application.

Claims

1. A performance testing method for an unmanned vehicle, characterized in that, include: The values ​​of control parameters within the performance testing program are calibrated. The performance testing program is pre-set in the vehicle control system of the autonomous vehicle to be tested, and the control parameters include braking trigger speed and target deceleration. A test start command is sent to the unmanned vehicle via remote control, so that the vehicle control system responds to the test start command and executes the test process defined in the performance test program according to the calibrated control parameters. The test process includes: controlling the unmanned vehicle to accelerate from a stationary state with a preset throttle opening, and when it is determined that the unmanned vehicle has accelerated to a speed that reaches the braking trigger speed, controlling the unmanned vehicle to stop accelerating and autonomously braking with a target deceleration until the unmanned vehicle comes to a stop. Acquire the driving data recorded by the autonomous vehicle during the execution of the test process; The acceleration and braking performance of the autonomous vehicle are analyzed based on the driving data.

2. The method according to claim 1, characterized in that, Before issuing a test start command to the driverless vehicle via remote control, the method further includes: The driverless vehicle is controlled to enter the autonomous driving ready state, and the driverless vehicle is ensured to be in the test initial state; the test initial state is: the driverless vehicle is in the parked state and the transmission is in neutral; Before controlling the autonomous vehicle to accelerate from a standstill at a preset throttle opening, the test procedure also includes: Control the unmanned vehicle to release the parking brake and switch the transmission to forward gear; In the test procedure, after confirming that the parking brake has been released and the transmission has been engaged in forward gear, the operation of controlling the driverless vehicle to accelerate from a stationary state with a preset throttle opening is then performed.

3. The method according to claim 2, characterized in that, After controlling the autonomous vehicle to stop accelerating and autonomously brake at a target deceleration, the test procedure also includes: After determining that the speed of the autonomous vehicle has dropped to no more than the preset parking speed and maintained for a preset observation time, the autonomous vehicle is controlled to enter the initial test state.

4. The method according to any one of claims 1-3, characterized in that, After controlling the autonomous vehicle to accelerate from a standstill at a preset throttle opening, the test procedure also includes: If the speed of the autonomous vehicle has not reached the braking trigger speed after accelerating for a preset waiting time, then when it is determined that the autonomous vehicle has accelerated for the preset waiting time, the operation of controlling the autonomous vehicle to stop accelerating and perform autonomous braking at the target deceleration is executed.

5. The method according to claim 1, characterized in that, After the test process is completed, the method further includes: A test stop command is issued to the unmanned vehicle via remote control to reset the performance test program configured in the vehicle control system to its initial state. The driverless vehicle can be remotely controlled to turn around or return to its original position. After the performance test program has been reset to its initial state and the autonomous vehicle has turned around or returned to its original position, the following first or second operation is performed: First operation: Send a test start command to the unmanned vehicle again via remote control, so that the vehicle control system responds to the resent test start command and executes the test process defined in the performance test program again according to the calibrated control parameters; The second operation is to recalibrate the values ​​of the control parameters in the performance test program; and to send a test start command to the unmanned vehicle again via remote control, so that the vehicle control system responds to the re-issued test start command and executes the test process defined in the performance test program according to the recalibrated control parameters.

6. The method according to claim 1, characterized in that, The performance testing program also defines emergency stop conditions, which include at least one of the following first, second, and third conditions: First condition: When the driverless vehicle is a new energy vehicle, the first condition is that the high-voltage system of the driverless vehicle is not powered on; when the driverless vehicle is a diesel-powered vehicle, the first condition is that the engine speed of the driverless vehicle is lower than the preset minimum execution speed. Second condition: The overall vehicle fault level of the unmanned vehicle is higher than level two; The third condition: The vehicle control system receives an emergency stop command issued via remote control. The step of issuing a test start command to the unmanned vehicle via remote control, so that the vehicle control system responds to the test start command and executes the test process defined in the performance test program according to the calibrated control parameters, includes: A test start command is sent to the autonomous vehicle via remote control, so that the vehicle control system responds to the test start command, executes the test process defined in the performance test program according to the calibrated control parameters, and performs a preset emergency braking operation on the autonomous vehicle when any of the emergency stop conditions is met during the execution of the test process.

7. The method according to claim 1 or 6, characterized in that, The test start command is issued by the remote controller; during the execution of the test process, the remote controller's control permissions over the unmanned vehicle are locked except for the emergency stop control permission; the emergency stop control permission is: the permission of the remote controller to issue an emergency stop command to the unmanned vehicle so that the vehicle control system can control the unmanned vehicle to perform a preset emergency braking operation.

8. The method according to claim 1, characterized in that, The step of analyzing the acceleration and braking performance of the autonomous vehicle based on the driving data includes: Based on the driving data of the autonomous vehicle during the acceleration phase, the acceleration performance index of the autonomous vehicle is determined; the acceleration phase is: controlling the autonomous vehicle to accelerate from a stationary state with a preset throttle opening until the autonomous vehicle accelerates to the point where the vehicle speed reaches the braking trigger speed; the acceleration performance index includes one or more of acceleration distance and acceleration time. Based on the driving data of the autonomous vehicle during the braking phase, the braking performance indicators of the autonomous vehicle are determined; the braking phase is: the process of controlling the autonomous vehicle to stop accelerating and autonomously braking at a target deceleration until the autonomous vehicle comes to a stop; the braking performance indicators include one or more of braking distance, braking duration and braking control accuracy.

9. The method according to claim 8, characterized in that, In the test process, the vehicle control system controls the autonomous vehicle to brake autonomously at a target deceleration. Specifically, it sets the braking force of the autonomous vehicle based on a preset benchmark calibration table to control the deceleration of the autonomous vehicle. The reference calibration table is used to provide the mapping relationship between braking force and the deceleration that can be achieved under the braking force. The braking performance indicators include braking control accuracy, and the braking control accuracy includes braking deceleration control accuracy. After determining the braking performance indicators of the autonomous vehicle, the method further includes: Based on the determined braking deceleration control accuracy, the mapping relationship given in the reference calibration table is updated.

10. A performance testing system for an unmanned vehicle, characterized in that, This includes driverless vehicles, remote controls, and test control and analysis equipment; The test control and analysis equipment is used to calibrate the values ​​of control parameters within the performance test program; the performance test program is preset in the vehicle control system of the unmanned vehicle, and the control parameters include braking trigger speed and target deceleration. The remote controller is used to issue a test start command to the driverless vehicle via remote control. The vehicle control system of the unmanned vehicle is used to respond to the test start command and execute the test process defined in the performance test program according to the calibrated control parameters. The test process includes: controlling the unmanned vehicle to accelerate from a stationary state with a preset throttle opening, and when it is determined that the unmanned vehicle has accelerated to a speed that reaches the braking trigger speed, controlling the unmanned vehicle to stop accelerating and autonomously braking with a target deceleration until the unmanned vehicle comes to a stop. The test control and analysis equipment is also used to acquire driving data recorded by the unmanned vehicle during the execution of the test process; The acceleration and braking performance of the autonomous vehicle are analyzed based on the driving data.

Images