Low-cost dynamic trigger test method and system for automatic driving multi-target vehicle conflict sequence in closed site

Abstract

The invention discloses a low-cost dynamic trigger test method and system for an automatic driving multi-target vehicle conflict sequence in a closed site, and belongs to the technical field of automatic driving test. The problem that it is difficult to verify the multi-vehicle conflict continuous processing capability of the autonomous vehicle VUT efficiently and at low cost through a single test is solved. The method is characterized by comprising the following steps: pre-defining a VUT route and a target vehicle VTi corresponding to a conflict point; obtaining a basic velocity-time curve of the VUT through preliminary operation; for each VTi, reversely calculating a pre-triggering position Ai according to a VUT dynamic speed curve and VTi calibration time ti, and calculating a distance L from the VUT to a conflict point and a time margin TTC in real time; when the VUT reaches Ai or the TTC is less than or equal to a preset threshold value Ti (including VTi response time and ti) of a corresponding conflict point, triggering a static VTi to start to make a conflict; and sequentially triggering each VTi to form a continuous conflict sequence. The method is used for verifying the multi-target conflict processing capability of the VUT at high efficiency and low cost in a single test.

Description

Technical Field

[0001] The present invention relates to the field of autonomous driving test verification technology. More specifically, the present invention relates to a low-cost dynamic triggering test method and system for autonomous driving multi-target vehicle conflict sequences in a closed field. Background Art

[0002] The rapid development of autonomous driving technology places extremely high demands on vehicle safety and reliability in various complex traffic scenarios. Closed-course testing is an important means of verifying the performance and safety margins of autonomous driving systems. Simulating common multi-vehicle interaction and conflict scenarios found on real roads, such as continuous intersection conflicts and multi-vehicle merging conflicts, is particularly critical. These scenarios effectively test the system's perception, decision-making, path planning, and emergency avoidance capabilities. However, efficiently, cost-effectively, and reliably constructing continuous, dynamic, multi-target vehicle conflict sequences in closed-course environments to fully cover the complex situations the system may encounter remains a significant challenge.

[0003] The current mainstream methods for testing multi-vehicle conflict scenarios in closed-loop test sites rely primarily on pre-set fixed trigger points or manually controlled target vehicles. These methods typically only allow for independent testing of a single or a small number of conflict points. Constructing a sequence of multiple conflict scenarios that occur sequentially often requires multiple independent test runs. This not only leads to low testing efficiency and extended testing cycles, but also significantly increases the site, equipment (such as the target vehicle), and personnel time costs, making it difficult to meet the rapid iteration requirements of development and testing.

[0004] Furthermore, the above method also shows the problem of insufficient dynamic adaptability. After the autonomous driving test vehicle encounters the first conflict target and takes evasive action (such as braking, steering), its speed trajectory will change significantly. If the subsequent conflict target vehicles are still triggered according to the preset fixed position or time, it is impossible to accurately simulate the dynamic process of the test vehicle in the real world encountering subsequent conflicts in the new state after evasion (such as reduced speed). This may lead to distortion of the test scenario, either the conflict is too loose (the test vehicle's speed has been reduced) or the conflict is too harsh (the trigger is too early and the test vehicle has not returned to the expected state), and it is impossible to accurately evaluate the system's ability to handle continuous dynamic conflicts.

[0005] Furthermore, even when dynamic triggering is attempted in existing technologies, accuracy and reliability issues are often encountered. For example, predicting the trigger point based on a single initial velocity curve cannot adapt to trajectory changes caused by the vehicle under test due to previous risk avoidance behavior; or relying solely on a single real-time collision time threshold to determine the trigger timing ignores the response time and movement time required for the target vehicle to start from a standstill. This may cause the target vehicle to be triggered too late and unable to accurately reach the predetermined conflict point, or to be triggered too early, resulting in invalid or unexpected conflict scenarios. These factors all limit the possibility of completing the verification of continuous multi-target conflict sequences efficiently, cost-effectively, and with high fidelity in a single test run. Therefore, there is an urgent need for a low-cost testing method that can dynamically, accurately, and efficiently trigger multi-target vehicle conflict sequences. Summary of the Invention

[0006] An object of the present invention is to solve at least the above problems and to provide at least the advantages which will be described hereinafter.

[0007] Another object of the present invention is to provide a low-cost dynamic triggering test method for the conflict sequence of multiple target vehicles in autonomous driving in a closed area. The method can complete the verification of the continuous conflict sequence of multiple target vehicles through a single test, significantly reducing costs and increasing efficiency; dynamically calculate the trigger point of subsequent target vehicles based on the actual speed of the VUT after risk avoidance, thereby improving the authenticity of the scene; combine the pre-trigger position (Ai) and the real-time collision time (TTC) dual threshold judgment and compensate the response time to ensure accurate triggering of the conflict.

[0008] To achieve these objectives and other advantages according to the present invention, a low-cost dynamic triggering test method for a closed-field autonomous driving multi-target vehicle conflict sequence is provided, comprising: S1. Predefine the driving route of the vehicle under test (VUT) and multiple conflict points in the closed test site. Each conflict point corresponds to an intervention position of the target vehicle (VTi), i = 1, 2, 3, etc., and execute the following data collection cycle: First test run: Control the VUT to drive the entire route without a target vehicle, and generate a basic speed-time curve for each location point; S2. For the i-th VTi, when i=1, based on the basic speed-time curve and the calibrated motion time ti of the VTi reaching the corresponding conflict point, the pre-trigger position Ai of the VUT at time ti before it reaches the conflict point is reversely calculated; when i≥2, after the VUT has gone through the i-1-th conflict point avoidance behavior, the actual speed-time curve is generated based on the actual speed and time data of its speed stabilization period, and the pre-trigger position Ai is recalculated; At the same time, the position and speed of the VUT are obtained in real time, and the real-time distance L and time margin TTC of the VUT to the conflict point are calculated in real time. TTC = L / VUT real-time speed; S3. When the VUT reaches the pre-trigger position Ai or the real-time TTC is less than or equal to the preset threshold Ti of the conflict point, a trigger command is sent to the corresponding target vehicle VTi. The VTi starts from rest and drives toward the conflict point along the preset trajectory, forming a preset conflict scenario with the VUT. S4: Dynamically trigger each VTi in sequence according to the VUT driving order to form a continuous conflict sequence, completing multi-scenario verification in a single test; Among them, the preset threshold Ti=VTi response time+ti; VTi response time is 50~200ms.

[0009] Preferably, in the process of recalculating the pre-trigger position Ai in step S2, a risk avoidance behavior intensity monitoring and compensation mechanism is implemented, specifically including: collecting the braking deceleration value of the VUT at the preceding conflict point in real time through the fusion of the vehicle-mounted inertial measurement unit and the wheel speed sensor, and when the measured deceleration value at any preceding conflict point exceeds the limit of 3.5m / s 2 , the compensation mechanism is activated: Calculate the arithmetic mean of all excessive decelerations in the preceding conflict points and generate the margin compensation value according to the following rules: 3.5m / s 2 ≤The arithmetic mean of all excessive decelerations <4.5m / s 2 When , the margin compensation value is 100ms; 4.5m / s 2 ≤The arithmetic mean of all excessive decelerations <5.5m / s 2 When , the margin compensation value is 150ms; The arithmetic mean of all excessive decelerations is >5.5m / s 2 When , the margin compensation value is 300ms; The generated margin compensation value is added to the preset threshold Ti of all subsequent conflict points, and the compensation remains effective until the end of the current test cycle.

[0010] Preferably, the method for obtaining the calibrated movement time ti for VTi to reach the corresponding conflict point in step S2 is as follows: Before the test, any target vehicle VTi is calibrated with no-load dynamics. Its average acceleration from rest to the conflict point is measured. Based on this acceleration and the measured distance between the VTi and the corresponding conflict point, the calibrated motion time of the VTi to the corresponding conflict point is calculated.

[0011] Preferably, when step S3 is executed, if the pre-trigger position condition and the real-time TTC condition are activated at the same time, the real-time TTC trigger instruction is forcibly selected as the effective trigger signal, and the other trigger channel is immediately locked until the current conflict scenario ends.

[0012] Preferably, a first control system is installed on the vehicle under test VUT, and the first control system includes an inertial navigation positioning module, a computing module and a first communication module arranged on the VUT, the inertial navigation positioning module includes at least one vehicle-mounted inertial measurement unit and a wheel speed sensor, and the computing module is communicatively connected to the inertial navigation positioning module and the first communication module. A second control system is installed on any target vehicle, and the second control system includes a throttle speed actuator, a VT control module and a second communication module, the VT control module is communicatively connected to the throttle speed actuator and the second communication module, and the first communication module is communicatively connected to the second communication module.

[0013] Preferably, the throttle speed actuator includes: An electric travel push rod, one end of which is rotatably connected to the body of the target vehicle; an accelerator pedal, one end of which is rotatably connected to the other end of the electric travel push rod, and the other end of which is rotatably connected to the body of the target vehicle; Wherein, the VT control module controls the stroke slope of the electric stroke push rod according to the trigger signal.

[0014] Preferably, the load rate of the computing module of the first control system is monitored in real time. When the load rate is continuously greater than 90% for 3 to 5 seconds, the real-time TTC trigger channel is automatically closed, a pure pre-trigger position trigger mode is adopted, and a hardware expansion alarm is generated and a degradation event is recorded.

[0015] Preferably, the low-cost dynamic triggering test method for the closed-field autonomous driving multi-target vehicle conflict sequence further includes establishing a multi-target vehicle trigger conflict arbitration mechanism, specifically performing the following steps: The pre-trigger position influence area associated with each conflict point is divided by electronic fence technology. The pre-trigger position influence area is a circular area with a radius of 4 to 6 meters and the pre-trigger position Ai as the center; Based on the current driving direction of the VUT, the target vehicles are given priority according to the following rules: Level 1: The target vehicle corresponding to the longitudinal conflict point that coincides with the extended line of the VUT trajectory; Level 2: target vehicle corresponding to the lateral conflict point orthogonal to the VUT trajectory; Level 3: Target vehicle corresponding to the oblique conflict point at an acute angle to the VUT trajectory; If it is detected that the pre-trigger position influence areas of multiple target vehicles overlap, the triggering order of the low-priority target vehicle will be forced to be postponed until the high-priority target vehicle completes the conflict; For the delayed triggered target vehicle, its calibrated motion time ti is reduced by 10% for each priority level delayed, and the compressed calibrated motion time shall not be less than 80% of the time required for the target vehicle to accelerate from rest to the corresponding conflict point at maximum acceleration, otherwise the triggering of the target vehicle will be cancelled.

[0016] The present invention further claims protection for a test system for the low-cost dynamic triggering test method for the closed-field autonomous driving multi-target vehicle conflict sequence, comprising: The first control system is installed on the vehicle under test (VUT), and includes: Inertial navigation positioning module, which includes at least a vehicle-mounted inertial measurement unit and a wheel speed sensor, and is used to collect VUT position, speed, and deceleration data in real time; A computation module that performs: Based on the predefined VUT driving route and conflict point sequence, a basic speed-time curve is generated during the first run without the target vehicle; When i=1, the pre-trigger position of the first conflict point is calculated based on the basic speed-time curve and the calibrated movement time ti of VTi reaching the corresponding conflict point. When i≥2, the pre-trigger position Ai of each target vehicle is dynamically calculated based on the actual speed-time curve after the previous conflict avoidance and the calibrated movement time ti of VTi reaching the corresponding conflict point. Calculate the real-time distance L and time margin TTC of the VUT to the i-th conflict point in real time; A first communication module, configured to send a trigger instruction; The second control system, which is installed on each target vehicle, includes: VT control module, used for receiving trigger instructions and generating control signals; a throttle speed actuator, which drives the target vehicle to move along a preset trajectory according to the control signal; a second communication module, which establishes a communication link with the first communication module; The dynamic trigger module is integrated into the calculation module and is used to: When the VUT reaches the pre-trigger position Ai or TTC ≤ the preset threshold Ti of the corresponding conflict point, a trigger command is sent to the corresponding target vehicle VTi; if the pre-trigger position and TTC conditions are activated at the same time, TTC trigger is forced to be selected and the other channel is locked; Among them, the calibrated motion time ti is obtained through no-load dynamic calibration, and the calibration method includes: measuring the average acceleration of the target vehicle VTi from standstill to the conflict point, and based on the acceleration and the measured distance between VTi and the corresponding conflict point, calculating the calibrated motion time for VTi to reach the corresponding conflict point.

[0017] Preferably, the calculation module further executes: Perform risk avoidance intensity monitoring: When the measured deceleration of the current conflict point exceeds the limit, a margin compensation value is generated and added to the preset threshold Ti of all subsequent conflict points; Load monitoring: If the computing module load rate remains > 90% for 3 to 5 seconds, the TTC trigger channel is closed, a hardware expansion alarm is generated, and a degradation event is recorded.

[0018] The present invention has at least the following beneficial effects: First, the present invention collects real-time speed and time data of the VUT after it has stabilized after avoiding a risk, and dynamically calculates the pre-trigger position Ai based on the actual speed-time curve. Combined with the dual trigger conditions of the real-time collision time TTC and the preset threshold Ti, a single test can accurately trigger a continuous collision sequence for multiple target vehicles, improving test efficiency while reducing site costs. Secondly, the present invention also monitors the braking deceleration limit of the preceding conflict point in real time and dynamically generates a margin compensation value, which is added to the threshold T of the subsequent conflict point. This effectively eliminates the cumulative deviation of the timing caused by strong risk avoidance behavior and ensures the spatiotemporal consistency of continuous conflict scenarios. Third, the present invention adopts a split control system architecture: the VUT side uses the inertial navigation positioning module and the calculation module to make real-time decisions on triggering instructions, and the target vehicle side responds and controls through the throttle speed actuator. The two systems interact synchronously through the communication module to ensure that the end-to-end delay from triggering instructions to vehicle execution is controlled within 200ms; Fourthly, the present invention also provides a calculation module load monitoring mechanism. When the load rate is >90% for 3-5 seconds, the real-time TTC calculation channel is automatically shut down, downgrading to pure pre-trigger position mode and generating a hardware expansion alarm to prevent system overload failure and improve the robustness of the test process. Fifth, the present invention further establishes a three-level conflict priority arbitration mechanism (longitudinal > horizontal > diagonal), forcibly delaying the triggering of low-priority target vehicles in the overlapping impact area, and dynamically compressing their calibrated motion time t, thereby ensuring the temporal authenticity of the core scene while avoiding scene failure caused by spatial conflicts.

[0019] Other advantages, objectives and features of the present invention will be reflected in part from the following description and will be understood by those skilled in the art through study and practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 A schematic diagram of a collision sequence of three target vehicles in a technical solution of the present invention; Figure 2 A schematic diagram of the first control system and the second control system in another technical solution of the present invention; Figure 3 It is a structural schematic diagram of the throttle speed actuator in another technical solution of the present invention. DETAILED DESCRIPTION

[0021] The present invention will be described in further detail below in conjunction with the accompanying drawings so that those skilled in the art can implement the invention with reference to the description.

[0022] It should be understood that terms such as “having”, “including” and “comprising” used herein do not preclude the existence or addition of one or more other elements or combinations thereof.

[0023] The present invention provides a low-cost dynamic triggering test method for a closed-field autonomous driving multi-target vehicle conflict sequence, comprising: S1. Predefine the driving route of the vehicle under test (VUT) and multiple conflict points in the closed test site. Each conflict point corresponds to an intervention position of the target vehicle (VTi), i = 1, 2, 3, etc., and execute the following data collection cycle: First test run: Control the VUT to drive the entire route without a target vehicle, and generate a basic speed-time curve for each location point; S2. For the i-th VTi, when i=1, based on the basic speed-time curve and the calibrated motion time ti of the VTi reaching the corresponding conflict point, the pre-trigger position Ai of the VUT at time ti before it reaches the conflict point is reversely calculated; when i≥2, after the VUT has gone through the i-1-th conflict point avoidance behavior, the actual speed-time curve is generated based on the actual speed and time data of its speed stabilization period, and the pre-trigger position Ai is recalculated; At the same time, the position and speed of the VUT are obtained in real time, and the real-time distance L and time margin TTC of the VUT to the conflict point are calculated in real time. TTC = L / VUT real-time speed; S3. When the VUT reaches the pre-trigger position Ai or the real-time TTC is less than or equal to the preset threshold Ti of the conflict point, a trigger command is sent to the corresponding target vehicle VTi. The VTi starts from rest and drives toward the conflict point along the preset trajectory, forming a preset conflict scenario with the VUT. S4: Dynamically trigger each VTi in sequence according to the VUT driving order to form a continuous conflict sequence, completing multi-scenario verification in a single test; Among them, the preset threshold Ti=VTi response time+ti; VTi response time is 50~200ms.

[0024] This technical solution enables low-cost testing of multi-target vehicle conflict sequences through dual trigger conditions and dynamic decision-making based on actual speed-time curves. First, a predefined route for the vehicle under test (VUT) and multiple conflict points are defined in a closed area, with each conflict point associated with a stationary target vehicle (VTi). A scenario model is constructed through data collection: a basic VUT speed-time curve is generated based on the speed-time data of the non-target vehicle.

[0025] For each target vehicle (VTi) (i=1, 2, 3, etc.), adaptive pre-trigger position calculation is employed. For the first target vehicle (i=1), the pre-trigger position Ai at time ti before the VUT reaches the conflict point is calculated in reverse based on the baseline speed-time curve and the VTi's calibrated motion time ti. For subsequent target vehicles (i≥2), Ai is recalculated based on the actual speed-time curve after the previous avoidance attempt, ensuring that the trigger point matches the VUT's real-time state. Simultaneously, real-time dual-channel monitoring is used to obtain the VUT's position and speed, and calculate its real-time distance L to the conflict point and its time margin TTC (TTC = L / real-time speed). Finally, dynamic triggering logic is employed: when the VUT reaches Ai or TTC ≤ the corresponding conflict point's preset threshold Ti (Ti = VTi response time + t, with a response time range of 50-200ms), the VTi is triggered to start from a standstill (response device: electromagnetic clutch + DC motor drive system), following a pre-set trajectory toward the conflict point, creating a precise conflict scenario.

[0026] like Figure 1 According to the above technical solution, taking the three target vehicle conflict sequence as an example, a specific workflow is as follows: Initialization and data collection: The VUT performs the first target-free run along the planned route, and generates a basic speed-time curve (sampling frequency 100 Hz) based on the recorded full-course data.

[0027] Dynamic trigger execution: VT1 triggering: Based on the base curve and VT1 calibration time t1, the pre-trigger position A1 (A' in the figure) is calculated. During real-time monitoring, VT1 is triggered when the VUT reaches A1 or TTC ≤ T (T = response time + t1).

[0028] VT2 triggering: A2 (B' in the figure) is recalculated based on the actual curve after VT1 avoidance. If the actual curve shows a decrease in VUT speed, A2 is moved forward compared to the calculated value on the base curve. The real-time TTC threshold is synchronously updated to T = response time + t2 value of VT2. After VT1 avoidance, the actual curve is generated by collecting at least six points after the VT1 conflict point is avoided and the speed stabilizes.

[0029] VT3 trigger: Similarly, A3 (C' in the figure) is calculated based on the actual curve after VT2 hedging to achieve time-series nested triggering.

[0030] Scenario closed loop: VT1, VT2, and VT3 are triggered in sequence according to the VUT driving order, and continuous verification of three scenarios is completed in a single test without interruption or repeated testing.

[0031] The above technical solution collects the speed curve after risk avoidance in real time and dynamically calculates the pre-trigger position based on this curve. Combined with real-time TTC dual-condition triggering, a single test can accurately generate a continuous conflict sequence for multiple target vehicles, compressing the scenario that traditionally requires N independent tests to one, improving test efficiency several times and reducing site costs. At the same time, it also improves the realism of the scenario: for subsequent target vehicles (i≥2), the pre-trigger position Ai is dynamically updated based on the actual speed curve after the previous risk avoidance in operation, ensuring that the conflict timing accurately matches the real-time motion state of the VUT (such as the trajectory after braking and deceleration), avoiding scene distortion caused by fixed trigger points. In addition, the dual-trigger condition design is compatible with vehicle positioning errors and response delays. The preset threshold Ti (including 50-200ms response time compensation) ensures the spatiotemporal accuracy of the conflict, and the trigger reliability reaches over 95%.

[0032] In one of the technical solutions, the risk avoidance behavior intensity monitoring and compensation mechanism is implemented during the recalculation of the pre-trigger position Ai in step S2, specifically including: collecting the braking deceleration value of the VUT at the preceding conflict point in real time through the fusion of the vehicle-mounted inertial measurement unit and the wheel speed sensor; when the measured deceleration value at any preceding conflict point exceeds the limit of 3.5m / s 2 , the compensation mechanism is activated: Calculate the arithmetic mean of all excessive decelerations in the preceding conflict points and generate the margin compensation value according to the following rules: 3.5m / s 2 ≤The arithmetic mean of all excessive decelerations <4.5m / s 2 When , the margin compensation value is 100ms; 4.5m / s 2 ≤The arithmetic mean of all excessive decelerations <5.5m / s 2 When , the margin compensation value is 150ms; The arithmetic mean of all excessive decelerations is >5.5m / s 2 When , the margin compensation value is 300ms; The generated margin compensation value is added to the preset threshold Ti of all subsequent conflict points, and the compensation remains effective until the end of the current test cycle.

[0033] The above technical solution further adds a risk avoidance behavior intensity monitoring and compensation mechanism, which uses on-board sensors to quantify the braking intensity of the VUT in the preceding conflict in real time and dynamically adjusts the subsequent trigger threshold to eliminate the accumulated deviation in timing. The specific implementation includes: Intensity monitoring: The actual braking deceleration value of each preceding conflict point is collected by integrating the inertial measurement unit (Bosch BMI088 can be selected for IMU) on the VUT vehicle and the wheel speed sensor (Hella SGM sensor can be selected). The value is set to 3.5m / s. 2The over-limit threshold is 3.5m / s 2 The over-limit threshold corresponds to medium emergency braking intensity and can be adjusted appropriately according to different test requirements.

[0034] Dynamic compensation: When the measured deceleration at any preceding conflict point exceeds 3.5m / s 2 , calculate the arithmetic mean of all over-limit decelerations and generate the margin compensation value according to the gradient rule: The average value is 3.5~4.5m / s 2 Time compensation 100ms (mild compensation); The average value is 4.5~5.5m / s 2 Time compensation 150ms (moderate compensation); The average value is >5.5m / s 2 The compensation time is 300ms (heavy compensation, corresponding to extreme braking).

[0035] Threshold correction: The compensation value is added to the preset threshold Ti of all subsequent conflict points (i.e., T' = original T + compensation value), and the compensation remains in effect until the end of the current test cycle.

[0036] According to the above technical solution, taking two consecutive conflict scenarios as an example (VT1→VT2), a specific implementation process is as follows: VT1 trigger and risk avoidance monitoring: After VT1 is triggered, the VUT brakes at the conflict point, and the deceleration value measured by the IMU and wheel speed sensor fusion exceeds 3.5m / s 2 Limit but less than 4.5m / s 2 .

[0037] The system activates the compensation mechanism: calculate the average value of the current over-limit deceleration and fall within 3.5~4.5m / s 2 interval, generating a 100ms margin compensation value.

[0038] VT2 trigger threshold correction: The original preset threshold T2 = VT2 response time + calibration movement time t2.

[0039] After adding the compensation value, the new threshold value T2' = T2 + 100ms. The preset threshold values ​​of all subsequent conflict points are added with 100ms.

[0040] Dynamic trigger execution: The VT2 pre-trigger position A2 is calculated based on the VT1 post-avoidance speed curve in the incremental learning of the conflict scenario (the VUT speed decreases due to braking, and the A2 position moves forward).

[0041] In real-time monitoring, when the VUT reaches A2 or TTC ≤ T2', VT2 is triggered, responding 100ms earlier than the original threshold to offset the VUT stroke shortening effect caused by the previous braking.

[0042] The above technical solution eliminates the timing interference of strong risk avoidance behavior, and through real-time monitoring of braking deceleration exceeding limit events and dynamic generation of compensation values, corrects the trigger threshold of subsequent conflict points, solves the problem of premature triggering of subsequent target vehicles due to sudden deceleration of VUT, and ensures the temporal and spatial consistency of continuous conflict scenarios. At the same time, the graded compensation mechanism accurately matches the impact of braking intensity: light braking (4.5m / s 2 Compensation within 100ms, heavy braking (> 5.5m / s 2 ) compensates for 300ms, ensuring a positive correlation between the compensation amount and the risk avoidance strength, thus avoiding over- or under-compensation. The compensation value persists until the end of the test, cumulatively correcting chain deviations between multiple conflict points and improving the fidelity of continuous scenarios.

[0043] In one technical solution, the method for obtaining the calibrated movement time ti of VTi reaching the corresponding conflict point in step S2 is as follows: Before testing, any target vehicle VTi undergoes no-load dynamic calibration, measuring its average acceleration from a standstill to the conflict point. Based on this acceleration and the measured distance between VTi and the corresponding conflict point, the calibrated motion time ti for VTi to reach the corresponding conflict point is calculated. By performing no-load dynamic calibration on the target vehicle VTi, measuring its average acceleration from a standstill to the conflict point, and accurately calculating the calibrated motion time ti based on the measured distance, this eliminates motion time deviations caused by individual performance differences of target vehicles (such as motor response and tire wear), ensures the spatiotemporal accuracy of the preset threshold Ti, keeps the conflict scenario triggering error within ±5%, and ensures the spatiotemporal consistency of the multi-target vehicle conflict sequence.

[0044] In one technical solution, during step S3, if both the pre-trigger position condition and the real-time time-to-collision condition are activated simultaneously, the real-time TTC trigger command is forcibly selected as the valid trigger signal, and the other trigger channel is immediately locked until the current conflict scenario ends. When both the pre-trigger position Ai and the real-time time-to-collision TTC condition are activated simultaneously, the TTC trigger command is forcibly selected and the other channel is locked, prioritizing the real-time collision risk (TTC ≤ T) and avoiding duplicate triggering or command conflicts caused by dual-channel competition. This locking mechanism also ensures the uniqueness of the trigger signal for the current conflict scenario, ensuring that the target vehicle VTi accurately intervenes according to the preset spatiotemporal conditions, and controlling the scenario construction error to within ±5%, significantly improving the trigger reliability of continuous conflict sequences.

[0045] like Figure 2In one technical solution, a first control system is installed on the vehicle under test (VUT). The first control system includes an inertial navigation and positioning module, a computing module, and a first communication module provided on the VUT. The inertial navigation and positioning module includes at least one vehicle-mounted inertial measurement unit and a wheel speed sensor. The computing module is communicatively connected to the inertial navigation and positioning module and the first communication module. A second control system is installed on any target vehicle. The second control system includes a throttle speed actuator, a VT control module, and a second communication module. The VT control module is communicatively connected to the throttle speed actuator and the second communication module, and the first communication module is communicatively connected to the second communication module.

[0046] The above technical solution implements the dynamic trigger logic through a split hardware architecture. The control component is split into a first control system deployed on the vehicle under test (VUT) and a second control system installed on the target vehicle. The first and second control systems work together via wireless communication. The first control system on the VUT includes an inertial positioning module, a computing module, and a first communication module. The inertial positioning module, which includes at least an onboard IMU and wheel speed sensors, collects real-time VUT position, speed, and deceleration data. It is mounted on the center of the VUT chassis and on all four wheel hubs. The computing module, which can be an NVIDIA Jetson AGX Xavier industrial computer, is installed in the VUT trunk and performs basic curve generation, pre-trigger position (Ai) calculation, real-time TTC monitoring, and trigger decision making. The first communication module, mounted on the VUT roof antenna box, is responsible for sending trigger commands to the target vehicle. The second control system on the VTi includes a VT control module, a throttle speed actuator, and a second communication module. The VT control module interprets trigger commands and generates motor control signals. The throttle speed actuator is mechanically connected to the accelerator pedal via a push rod. The second communication module receives commands from the first control system.

[0047] This technical solution not only ensures the reliability of the trigger signal, but also isolates the VUT from the target vehicle control risks through a split architecture. The VUT focuses on perception and decision-making (computing module), while the target vehicle focuses on execution (motor mechanism), preventing single-point failures from causing system crashes. It also reduces end-to-end latency, compressing the entire chain of command transmission, parsing, and execution to within 200ms, meeting the 50-200ms response time requirement. Furthermore, the target vehicle only needs to be equipped with a universal actuator, supporting rapid deployment of multiple target vehicle sequences and reducing hardware costs.

[0048] like Figure 3 As shown, in one of the technical solutions, the throttle speed actuator includes: An electric travel push rod 2, one end of which is rotatably connected to the target vehicle body 1; an accelerator pedal 3, one end of which is rotatably connected to the other end of the electric travel push rod, and the other end of which is rotatably connected to the target vehicle body 1; The VT control module controls the stroke slope of the electric stroke push rod 2 according to the trigger signal.

[0049] The above technical solution discloses an optimal mechanical structure of a throttle speed execution structure, wherein the throttle speed execution mechanism realizes precise control of acceleration through a dual-degree-of-freedom rotation structure, and includes an electric stroke push rod 2 connected to a target vehicle body 1 through a first rotating body 4 and an accelerator pedal 3 connected to the electric stroke push rod 2 through a second rotating body 5. The first rotating body 4 and the second rotating body 5 are preferably pin-hole bodies. The assembly position of the electric stroke push rod 2 is the front firewall of the engine compartment, and the assembly position of the accelerator pedal 3 is the cockpit floor. The VT control module maps the target acceleration required by the trigger signal to the stroke slope of the electric stroke push rod 2, and adjusts the speed of the electric stroke push rod motor through a PWM signal.

[0050] This technical solution achieves linear controllable acceleration. The dual-rotation structure eliminates side loads on the push rod, fully converting thrust into pedal torque. Combined with precise slope control, it ensures that the target vehicle moves according to the preset acceleration curve. Furthermore, the use of standardized electric stroke push rods instead of servo hydraulic systems significantly reduces hardware costs.

[0051] One technical solution monitors the load rate of the computing module of the primary control system in real time. If the load rate exceeds 90% for 3 to 5 seconds, the real-time TTC trigger channel is automatically shut down, the pre-trigger position trigger mode is adopted, a hardware expansion alarm is generated, and a degradation event is recorded. By monitoring the computing module load rate in real time, the real-time time to collision (TTC) trigger channel is automatically shut down and the system is downgraded to the pre-trigger position trigger mode, preventing system overload and crashes. Simultaneously generating a hardware expansion alarm and recording a degradation event ensures test continuity while providing a basis for operational and maintenance decision-making, thereby improving system availability.

[0052] In one technical solution, the low-cost dynamic triggering test method for a closed-field autonomous driving multi-target vehicle conflict sequence further includes establishing a multi-target vehicle trigger conflict arbitration mechanism, specifically performing the following steps: The pre-trigger position influence area associated with each conflict point is divided by electronic fence technology. The pre-trigger position influence area is a circular area with a radius of 4 to 6 meters and the pre-trigger position Ai as the center; Based on the current driving direction of the VUT, the target vehicles are given priority according to the following rules: Level 1: The target vehicle corresponding to the longitudinal conflict point that coincides with the extended line of the VUT trajectory; Level 2: target vehicle corresponding to the lateral conflict point orthogonal to the VUT trajectory; Level 3: Target vehicle corresponding to the oblique conflict point at an acute angle to the VUT trajectory; If it is detected that the pre-trigger position influence areas of multiple target vehicles overlap, the triggering order of the low-priority target vehicle will be forced to be postponed until the high-priority target vehicle completes the conflict; For the delayed triggered target vehicle, its calibrated motion time ti is reduced by 10% for each priority level delayed, and the compressed calibrated motion time shall not be less than 80% of the time required for the target vehicle to accelerate from rest to the corresponding conflict point at maximum acceleration, otherwise the triggering of the target vehicle will be cancelled.

[0053] The above technical solution solves the conflict problem caused by overlapping trigger areas of multiple target vehicles through a spatial priority arbitration mechanism. Through electronic fence partitioning, a circular influence area with a radius of 4 to 6 meters is constructed with each target vehicle's pre-trigger position (Ai) as the center, forming the pre-trigger position influence area. At the same time, three levels of dynamic priority are set: Level 1: Longitudinal conflict target vehicle (coinciding with the VUT trajectory extension line, such as rear-end collision scenario), given the highest priority; Level 2: Lateral conflict target vehicle (orthogonal to the VUT trajectory, such as cutting into an intersection), medium priority; Level 3: oblique conflict target vehicle (forming an acute angle with the trajectory, such as merging area conflict), lowest priority.

[0054] When multiple overlapping impact zones are detected, the triggering of the low-priority target vehicle is forcibly delayed until the high-priority target vehicle completes the conflict. When the triggering is delayed, the calibrated motion time ti is reduced by 10% for each delay, and after compression, ti ≥ 80% of the time required for maximum acceleration (lower limit protection).

[0055] This technical solution ensures the temporal authenticity of high-value scenarios. By precisely quantifying impact zone overlap through geo-fencing (radius 4-6m), combined with three-level priority-based mandatory scheduling, it avoids the risk of physical collisions caused by simultaneous triggering of multiple target vehicles, thus reducing the accident rate. A TI compression mechanism for delayed triggering of target vehicles compensates for time delays, and combined with a lower limit protection, strictly controls the temporal and spatial errors of delayed scenarios, ensuring test effectiveness. Prioritization focuses on core longitudinal and lateral scenarios (representing 80% of real-world accidents) to avoid diagonal conflicts and interference. A single test can handle multiple overlapping target vehicles, further improving site utilization.

[0056] The present invention further claims protection for a test system for the low-cost dynamic triggering test method for the closed-field autonomous driving multi-target vehicle conflict sequence, comprising: The first control system is installed on the vehicle under test (VUT), and includes: Inertial navigation positioning module, which includes at least a vehicle-mounted inertial measurement unit and a wheel speed sensor, and is used to collect VUT position, speed, and deceleration data in real time; A computation module that performs: Based on the predefined VUT driving route and conflict point sequence, a basic speed-time curve is generated during the first run without the target vehicle; When i=1, the pre-trigger position of the first conflict point is calculated based on the basic speed-time curve and the calibrated movement time ti of VTi reaching the corresponding conflict point. When i≥2, the pre-trigger position Ai of each target vehicle is dynamically calculated based on the actual speed-time curve after the previous conflict avoidance and the calibrated movement time ti of VTi reaching the corresponding conflict point. Calculate the real-time distance L and time margin TTC of the VUT to the i-th conflict point in real time; A first communication module, configured to send a trigger instruction; The second control system, which is installed on each target vehicle, includes: VT control module, used for receiving trigger instructions and generating control signals; a throttle speed actuator, which drives the target vehicle to move along a preset trajectory according to the control signal; a second communication module, which establishes a communication link with the first communication module; The dynamic trigger module is integrated into the calculation module and is used to: When the VUT reaches the pre-trigger position Ai or TTC ≤ the preset threshold Ti of the corresponding conflict point, a trigger command is sent to the corresponding target vehicle VTi; if the pre-trigger position and TTC conditions are activated at the same time, TTC trigger is forced to be selected and the other channel is locked; Among them, the calibrated motion time ti is obtained through no-load dynamic calibration, and the calibration method includes: measuring the average acceleration of the target vehicle VTi from standstill to the conflict point, and based on the acceleration and the measured distance between VTi and the corresponding conflict point, calculating the calibrated motion time for VTi to reach the corresponding conflict point.

[0057] The present invention further claims protection for a test system based on a low-cost trigger test method for multi-target conflict sequences in closed-field autonomous driving. This system achieves full closed-loop control through an integrated system architecture. The inertial navigation positioning module and the computing module of the first control system work together to accurately generate basic curves, dynamically calculate pre-trigger positions, and monitor real-time time-to-collision (TTC). The throttle speed actuator and the VT control module of the second control system ensure that the target vehicle responds precisely to commands. The two systems interact in real time via a communication link, compressing end-to-end latency to within 200ms and ensuring synchronization of the triggering timing of multiple target vehicles. The dynamic trigger module utilizes a dual-condition decision mechanism (pre-trigger position Ai or TTC ≤ T) and forces TTC triggering to take precedence. Combined with a channel locking function, this system reliably constructs a continuous conflict sequence within a single test, reducing the system's false trigger rate to below 1%, significantly improving test efficiency.

[0058] In one of the technical solutions, the calculation module further performs: Perform risk avoidance intensity monitoring: When the measured deceleration of the current conflict point exceeds the limit, a margin compensation value is generated and added to the preset threshold Ti of all subsequent conflict points; Load monitoring: If the computing module load rate remains > 90% for 3 to 5 seconds, the TTC trigger channel is closed, a hardware expansion alarm is generated, and a degradation event is recorded.

[0059] In the above technical solution, the test system can detect in real time the over-limit of the measured deceleration at the preceding conflict point and automatically generate a margin compensation value through the risk avoidance intensity monitoring mechanism of the calculation module, and add it to the preset threshold Ti of all subsequent conflict points, thereby dynamically compensating for the speed deviation caused by the VUT risk avoidance behavior, significantly improving the accuracy of the continuous conflict sequence triggering and the authenticity of the test scenario, and avoiding the problem of false triggering or missed triggering of subsequent conflict points due to excessive risk avoidance intensity. In addition, through the load monitoring function, the system automatically closes the real-time TTC trigger channel and downgrades to a pure pre-trigger position mode when the load rate of the calculation module continues to exceed the limit. At the same time, it generates a hardware expansion alarm and records the degradation event, effectively preventing calculation delays or trigger failures caused by system overload, ensuring the stability and reliability of the test process, and providing data support for hardware optimization to enhance overall robustness.

[0060] Although the embodiments of the present invention have been disclosed above, they are not limited to the applications listed in the description and implementation methods. They can be fully applied to various fields suitable for the present invention. For those familiar with the art, additional modifications can be easily implemented. Therefore, without departing from the general concept defined by the claims and the scope of equivalents, the present invention is not limited to the specific details and illustrations shown and described herein.

Claims

1. A low-cost dynamic triggering test method for a multi-target vehicle conflict sequence in an autonomous driving environment in a closed field, characterized by: include: S1. Predefine the driving route of the vehicle under test (VUT) and multiple conflict points in the closed test site. Each conflict point corresponds to an intervention position of the target vehicle (VTi), i = 1, 2, 3, etc., and execute the following data collection cycle: First test run: Control the VUT to drive the entire route without a target vehicle, and generate a basic speed-time curve for each location point; S2. For the i-th VTi, when i=1, based on the basic speed-time curve and the calibrated motion time ti of the VTi reaching the corresponding conflict point, the pre-trigger position Ai of the VUT at time ti before it reaches the conflict point is reversely calculated; when i≥2, after the VUT has gone through the i-1-th conflict point avoidance behavior, the actual speed-time curve is generated based on the actual speed and time data of its speed stabilization period, and the pre-trigger position Ai is recalculated; At the same time, the position and speed of the VUT are obtained in real time, and the real-time distance L and time margin TTC of the VUT to the conflict point are calculated in real time. TTC = L / VUT real-time speed; S3. When the VUT reaches the pre-trigger position Ai or the real-time TTC is less than or equal to the preset threshold Ti of the conflict point, a trigger command is sent to the corresponding target vehicle VTi. The VTi starts from rest and drives toward the conflict point along the preset trajectory, forming a preset conflict scenario with the VUT. S4: Dynamically trigger each VTi in sequence according to the VUT driving order to form a continuous conflict sequence, completing multi-scenario verification in a single test; Among them, the preset threshold Ti=VTi response time+ti; VTi response time is 50~200ms.

2. The low-cost dynamic triggering test method for a closed-field autonomous driving multi-target vehicle conflict sequence according to claim 1, characterized in that: In the process of recalculating the pre-trigger position Ai in step S2, the risk avoidance behavior intensity monitoring and compensation mechanism is implemented, specifically including: collecting the braking deceleration value of the VUT at the previous conflict point in real time through the fusion of the on-board inertial measurement unit and the wheel speed sensor, and when the measured deceleration value of any previous conflict point exceeds the limit of 3.5m / s 2 , the compensation mechanism is activated: Calculate the arithmetic mean of all excessive decelerations in the preceding conflict points and generate the margin compensation value according to the following rules: 3.5m / s 2 ≤The arithmetic mean of all excessive decelerations <4.5m / s 2 When , the margin compensation value is 100ms; 4.5m / s 2 ≤The arithmetic mean of all excessive decelerations <5.5m / s 2 When , the margin compensation value is 150ms; The arithmetic mean of all excessive decelerations is >5.5m / s 2 When , the margin compensation value is 300ms; The generated margin compensation value is added to the preset threshold Ti of all subsequent conflict points, and the compensation remains effective until the end of the current test cycle.

3. The low-cost dynamic triggering test method for a closed-field autonomous driving multi-target vehicle conflict sequence according to claim 2, characterized in that: The method for obtaining the calibrated motion time ti for VTi to reach the corresponding conflict point in step S2 is as follows: Before the test, any target vehicle VTi is calibrated with no-load dynamics. Its average acceleration from rest to the conflict point is measured. Based on this acceleration and the measured distance between the VTi and the corresponding conflict point, the calibrated motion time of the VTi to the corresponding conflict point is calculated.

4. The low-cost dynamic triggering test method for a closed-field autonomous driving multi-target vehicle conflict sequence according to claim 3, characterized in that: When step S3 is executed, if the pre-trigger position condition and the real-time TTC condition are activated at the same time, the real-time TTC trigger instruction is forcibly selected as the effective trigger signal, and the other trigger channel is immediately locked until the current conflict scenario ends.

5. The low-cost dynamic triggering test method for a closed-field autonomous driving multi-target vehicle conflict sequence according to claim 4, characterized in that: A first control system is installed on the vehicle under test VUT, and the first control system includes an inertial navigation positioning module, a computing module and a first communication module arranged on the VUT. The inertial navigation positioning module includes at least one vehicle-mounted inertial measurement unit and a wheel speed sensor. The computing module is communicatively connected to the inertial navigation positioning module and the first communication module. A second control system is installed on any target vehicle, and the second control system includes a throttle speed actuator, a VT control module and a second communication module. The VT control module is communicatively connected to the throttle speed actuator and the second communication module, and the first communication module is communicatively connected to the second communication module.

6. The low-cost dynamic triggering test method for a closed-field autonomous driving multi-target vehicle conflict sequence according to claim 5, characterized in that: The throttle speed actuator comprises: An electric travel push rod, one end of which is rotatably connected to the body of the target vehicle; an accelerator pedal, one end of which is rotatably connected to the other end of the electric travel push rod, and the other end of which is rotatably connected to the body of the target vehicle; Wherein, the VT control module controls the stroke slope of the electric stroke push rod according to the trigger signal.

7. The low-cost dynamic triggering test method for a closed-field autonomous driving multi-target vehicle conflict sequence according to claim 6, characterized in that: Monitor the load rate of the computing module of the first control system in real time. When the load rate is >90% for 3 to 5 seconds, the real-time TTC trigger channel is automatically closed, the pure pre-trigger position trigger mode is adopted, and a hardware expansion alarm is generated and a degradation event is recorded.

8. The low-cost dynamic triggering test method for a closed-field autonomous driving multi-target vehicle conflict sequence according to claim 7, characterized in that: It also includes establishing a multi-target vehicle-triggered conflict arbitration mechanism, specifically performing the following steps: The pre-trigger position influence area associated with each conflict point is divided by electronic fence technology. The pre-trigger position influence area is a circular area with a radius of 4 to 6 meters and the pre-trigger position Ai as the center; Based on the current driving direction of the VUT, the target vehicles are given priority according to the following rules: Level 1: The target vehicle corresponding to the longitudinal conflict point that coincides with the extended line of the VUT trajectory; Level 2: target vehicle corresponding to the lateral conflict point orthogonal to the VUT trajectory; Level 3: Target vehicle corresponding to the oblique conflict point at an acute angle to the VUT trajectory; If it is detected that the pre-trigger position influence areas of multiple target vehicles overlap, the triggering order of the low-priority target vehicle will be forced to be postponed until the high-priority target vehicle completes the conflict; For the delayed triggered target vehicle, its calibrated motion time ti is reduced by 10% for each priority level delayed, and the compressed calibrated motion time shall not be less than 80% of the time required for the target vehicle to accelerate from rest to the corresponding conflict point at maximum acceleration, otherwise the triggering of the target vehicle will be cancelled.

9. A test system based on the low-cost dynamic triggering test method for the closed-field autonomous driving multi-target vehicle conflict sequence according to any one of claims 1 to 8, characterized in that: include: The first control system is installed on the vehicle under test (VUT), and includes: Inertial navigation positioning module, which includes at least a vehicle-mounted inertial measurement unit and a wheel speed sensor, and is used to collect VUT position, speed, and deceleration data in real time; A computation module that performs: Based on the predefined VUT driving route and conflict point sequence, a basic speed-time curve is generated during the first run without the target vehicle; When i=1, the pre-trigger position of the first conflict point is calculated based on the basic speed-time curve and the calibrated movement time ti of VTi reaching the corresponding conflict point. When i≥2, the pre-trigger position Ai of each target vehicle is dynamically calculated based on the actual speed-time curve after the previous conflict avoidance and the calibrated movement time ti of VTi reaching the corresponding conflict point. Calculate the real-time distance L and time margin TTC of the VUT to the i-th conflict point in real time; A first communication module, configured to send a trigger instruction; The second control system, which is installed on each target vehicle, includes: VT control module, used for receiving trigger instructions and generating control signals; a throttle speed actuator, which drives the target vehicle to move along a preset trajectory according to the control signal; a second communication module, which establishes a communication link with the first communication module; The dynamic trigger module is integrated into the calculation module and is used to: When the VUT reaches the pre-trigger position Ai or TTC ≤ the preset threshold Ti of the corresponding conflict point, a trigger command is sent to the corresponding target vehicle VTi; if the pre-trigger position and TTC conditions are activated at the same time, TTC trigger is forced to be selected and the other channel is locked; Among them, the calibrated motion time ti is obtained through no-load dynamic calibration, and the calibration method includes: measuring the average acceleration of the target vehicle VTi from standstill to the conflict point, and based on the acceleration and the measured distance between VTi and the corresponding conflict point, calculating the calibrated motion time for VTi to reach the corresponding conflict point.

10. The test system according to claim 9, wherein: The calculation module also performs: Perform risk avoidance intensity monitoring: When the measured deceleration of the current conflict point exceeds the limit, a margin compensation value is generated and added to the preset threshold Ti of all subsequent conflict points; Load monitoring: If the computing module load rate remains > 90% for 3 to 5 seconds, the TTC trigger channel is closed, a hardware expansion alarm is generated, and a degradation event is recorded.

Images