Simulation test evaluation method and evaluation system for vehicle emergency lane keeping system

Abstract

This invention provides a simulation test and evaluation method and system for a vehicle emergency lane keeping system. By testing the ELK system software in a large number of different scenarios in a simulation environment, the system's performance is fully verified, effectively evaluating the ELK system's functionality. Rapid testing during function development allows for early identification of algorithmic logic issues. System verification testing is performed later in development to confirm whether design objectives have been achieved. This significantly shortens the actual vehicle testing, verification, and calibration testing cycles, reducing development and testing costs. The invention supports virtual parameter calibration and provides system optimization recommendations.

Description

Technical Field

[0001] The present invention belongs to the technical field of simulation testing of advanced driver assistance systems, and in particular relates to a simulation test evaluation method and an evaluation system for a vehicle emergency lane keeping system. Background Art

[0002] Advanced driver assistance systems (ADAS)

[0003] Emergency Lane Keeping System (ELK)

[0004] Autonomous Emergency Braking System (AEB)

[0005] Model-in-the-Loop (MiL)

[0006] Hardware-in-the-Loop (HiL)

[0007] Time to line crossing (TLC)

[0008] This technology belongs to the field of advanced driver assistance system simulation testing. Within this field, there is a lack of patents related to ADAS simulation testing and evaluation, particularly simulation testing methods and evaluation systems for ELK. Currently, there are only patents for simulation testing methods for AEB, but they do not cover system evaluation methods and systems. Other ELK patents also focus on ELK function implementation methods and do not cover simulation testing and system performance evaluation.

[0009] The Emergency Lane Keeping System (ELK) is usually an ADAS system that is turned on by default in a vehicle. When the vehicle deviates from its current lane, if the lane line is not a dotted line, or there is a curb on the side of the deviated lane, or the lane line is a dotted line but there is a risk of collision with other traffic participants, the ELK system will intervene and issue an emergency intervention to the vehicle's steering system to keep the vehicle in its original lane.

[0010] Because the operating conditions when this function is triggered are relatively dangerous (the TLC value is low), and the test vehicle's lateral acceleration and rate of change are large when the system intervenes, actual vehicle performance testing is risky and prone to understeer, oversteer, and even loss of control. Therefore, comprehensive and safe testing of the ELK function during the development and testing phases is crucial. To this end, a simulation-based evaluation system is urgently needed. Summary of the Invention

[0011] The technical problem to be solved by the present invention is to provide a simulation test evaluation method and evaluation system for a vehicle emergency lane keeping system, which are used to evaluate the emergency lane keeping system ELK.

[0012] The technical solution adopted by the present invention to solve the above technical problems is: a simulation test and evaluation method for a vehicle emergency lane keeping system, comprising the following steps:

[0013] S0: Build a simulation test and evaluation system for a hardware-in-the-loop (HiL) environment, including a host computer, a slave computer, and the ADAS domain controller under test. The host computer runs test management software, and the host and slave computers communicate via Ethernet. The slave computers run Carmaker scenario simulation software, vehicle dynamics models, signal transmission models, and driver boards to implement simulation scenario operation, vehicle dynamics response, CAN communication with the ECU under test, and real-time test data feedback to the host computer. The ADAS domain controller runs the test object ELK.

[0014] S1: Establish ELK test scenarios in the simulation software according to the ELK function definition document and test requirement document;

[0015] S2: Evaluate the functions and performance of the ELK system and give specific scores for each simulation scenario in the ELK system to determine whether it meets the system requirements.

[0016] According to the above scheme, in step S0, the specific steps are:

[0017] In the ELK simulation test, the key data of the test vehicle, the key data of the target vehicle obtained by the sensor model in the simulation environment, and the lane line data are transmitted to the ADAS domain controller via the CAN network;

[0018] The controller's integrated ELK application software uses this data to comprehensively determine whether to intervene and outputs the steering angle control value when intervening. The control value is then fed back to the vehicle dynamics software and Carmaker scenario software to achieve closed-loop control of the test vehicle's dynamics.

[0019] Key data of the test vehicle include ELK function switch signal, vehicle speed and gear position;

[0020] The key data of the target vehicle obtained by the sensor model in the simulation environment include the lane where the target vehicle is located, relative distance, relative speed and relative acceleration;

[0021] Lane line data includes lane line type and lane line coefficient.

[0022] According to the above scheme, in step S1, the specific steps are:

[0023] S11: Simulate the vehicle power-on operation, and monitor the ELK status bit to see if it meets the function definition after power-on is completed;

[0024] S12: The model ELK function is turned on and the ELK status is monitored to see if it is correctly switched. During this stage, the driver model in the simulation software controls the vehicle in both the lateral and longitudinal directions. This control includes steering, braking, and driving.

[0025] S13: The driver model is set in the scene to input a steering angle to the control vehicle, and the test vehicle begins to deviate from the center line of the original lane;

[0026] Determine whether the monitoring ELK issues an alarm. If an alarm is issued, output the distance to the lane line and the TLC value at the time of the alarm;

[0027] If ELK does not intervene during the deviation process, the simulation will be stopped when the vehicle completely deviates from the original lane and the test data will be recorded;

[0028] If the ELK issues a steering intervention request during the deviation process, the test vehicle is immediately switched from driver model control to ADAS controller control, and the vehicle dynamics model responds to the angle intervention request value issued by the ELK to implement the ELK steering intervention function;

[0029] During ELK intervention, the ELK steering angle request value is recorded in real time, and the test vehicle's lateral velocity, acceleration, jerk, yaw rate, yaw acceleration, and TLC values ​​are recorded until the ELK intervention is terminated or the lane deviates from the lane.

[0030] According to the above scheme, in step S2, the single scenario evaluation includes functional indicators and performance indicators; if all functional indicators are met, 1 point is recorded, and if they are not met, 0 point is recorded; the performance indicators are scored according to the achievement;

[0031] Assume that the evaluation result of the functional index of the ELK system is fi, specifically:

[0032] f1 is used to monitor whether the ELK function state machine is consistent with the function definition during the entire simulation process, including waiting, activation, inhibition and fault states;

[0033] f2 means that when the lane line is dotted, there is an adjacent lane, and there is no collision risk, the ELK function will not interfere with the test vehicle's lane change and will not issue an alarm;

[0034] f3 means that when the lane line is not a dotted line, there is an adjacent lane, the driver has turned on the turn signal and there is no collision risk, the driver is considered to have an active lane change intention. The ELK function does not interfere with the test vehicle's lane change and does not issue an alarm;

[0035] f4 refers to the typical ELK operating condition when the lane line is not a dotted line, there is an adjacent lane, and the turn signal is not on. An alarm is issued and the test vehicle is intervened to change lanes;

[0036] f5 refers to the situation where there is a dashed lane line, an adjacent lane exists, and the test vehicle deviates from the original lane, resulting in a collision risk. This is a typical ELK operating condition, in which an alarm is issued and the test vehicle is intervened to change lanes;

[0037] f6 refers to the situation where there is a dashed lane line, an adjacent lane exists, and the test vehicle deviates from the original lane, which poses a collision risk. This is a typical ELK operating condition, in which an alarm is issued and the test vehicle is intervened to change lanes;

[0038] f7 means the test vehicle deviates from the curb, which is a typical ELK operating condition. An alarm is issued and the test vehicle is intervened to change lanes;

[0039] f8 is the basic requirement of the ELK function. During the ELK intervention process, the test vehicle must not deviate from the original lane and the crossing line must be less than 30cm;

[0040] Based on the ELK function definition and driving safety requirements, f9 constrains the test vehicle's heading angle after ELK intervention, ensuring that the test vehicle does not deviate from its original lane within 50 meters after ELK intervention.

[0041] Furthermore, in step S2, the function achievement score F is introduced to evaluate the functional indicators of a certain simulation scenario test of ELK:

[0042]

[0043] If and only if f i When (i=1, 2, 3, 4, ...) are all 1 and all functional indicators of ELK are met, F=1;

[0044] The specific performance indicators of the ELK system are:

[0045] TLC warning Used to calculate the timing of the test vehicle deviation alarm issued by the ELK system and to evaluate whether the alarm timing is too early or too late;

[0046] D warning Used to evaluate the time when the test vehicle's departure warning is issued by the ELK system and the distance from the lane line, and to evaluate whether the warning distance is too early or too late;

[0047] TLC steering Used to calculate the timing of the ELK system's steering intervention and to evaluate whether the steering intervention is too early or too late;

[0048] D steering Used to assess the distance between the test vehicle and the lane line when the ELK system intervenes in steering, and to evaluate whether the steering intervention distance is too early or too late;

[0049] v y Refers to the simulated lateral speed of the test vehicle and the assessment of its absolute value;

[0050] a y Refers to the lateral acceleration of the test vehicle, not exceeding 6m / s^2;

[0051] a″ y Refers to the rate of change of the lateral acceleration of the test vehicle, which shall not exceed 8m / s^3;

[0052] Yaw rate Refers to the yaw angular velocity of the test vehicle during simulation;

[0053] Yaw acc Refers to the rate of change of yaw angular velocity during the simulation of the test vehicle;

[0054] LatDisplace min Used to calculate the effect of ELK intervention on the vehicle's lateral movement. A non-negative value means that the vehicle is not allowed to cross the line and a safe lateral distance of 10 cm is left.

[0055] Yaw terminate Used to calculate the yaw angle of the test vehicle relative to the center line of the original lane after ELK intervention is completed.

[0056] Furthermore, in step S2, the data of each of the above performance indicators are evaluated based on the test purpose and test content of each simulation scenario, with a full score of 10 points; factors affecting the scoring result include driving safety, driving comfort and user complaints;

[0057] Driving safety means that ELK intervention should not be too late, based on the complexity and risk of the scene, to avoid the risk of vehicles crossing the line;

[0058] Driving comfort means that the maximum values ​​of vehicle acceleration and acceleration change rate must be within a reasonable range while avoiding collision risks.

[0059] User complaint level means that when there is a certain risk, the ELK system should not intervene too early to avoid frequent intervention and steering that may cause user complaints.

[0060] Furthermore, in step S2, the weight of each performance indicator score is calculated to obtain the performance evaluation score of the ELK system in a certain simulation scenario; the performance indicator weight of the ELK system is comprehensively obtained based on the dimensions including driving safety, driving comfort and user complaints:

[0061] TLC steering It is an important indicator for judging whether the system's intervention moment is appropriate and is given the highest weight;

[0062] LatDisplace min It is an important indicator for judging the effectiveness of the system's steering intervention. It is used to calculate the minimum lateral distance between the vehicle and the lane line and to determine whether the vehicle has crossed the lane line. It is given the second highest weight.

[0063] This yields the weight values ​​corresponding to each performance evaluation parameter:

[0064] ω i (∑ω i =1,ω i >0; i=1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11,...),

[0065] The performance evaluation of ELK in a certain scenario is P:

[0066]

[0067] The comprehensive evaluation of ELK in a certain scenario is S:

[0068] S=F*P

[0069] S is the comprehensive evaluation score, which requires that all functional indicators are met, and the score is determined by the performance evaluation indicators.

[0070] The simulation test and evaluation system for the vehicle's emergency lane keeping system uses a hardware-in-the-loop (HiL) approach and includes a host computer, a slave computer, and an ADAS domain controller under test. The host computer runs test management software to invoke simulation test scenarios and control system models, and the host and slave computers communicate via Ethernet. The slave computers run Carmaker scenario simulation software, vehicle dynamics models, signal transmission models, and driver boards to implement simulation scenario operation, vehicle dynamics response, CAN communication with the ECU under test, and real-time test data feedback to the host computer. The ADAS domain controller runs the test object ELK to issue control commands.

[0071] A computer storage medium stores a computer program that can be executed by a computer processor. The computer program executes a simulation test and evaluation method for a vehicle emergency lane keeping system.

[0072] The beneficial effects of the present invention are:

[0073] 1. The simulation test evaluation method and evaluation system of the vehicle emergency lane keeping system of the present invention fully verifies the performance of the ELK system by testing the ELK system software in a large number of different scenarios in a simulation environment, and realizes the function of evaluating the emergency lane keeping system ELK.

[0074] 2. The present invention conducts rapid testing during function development to discover functional algorithm logic problems early; performs system verification testing in the later stages of development to confirm whether the design goals have been achieved; and significantly shortens the actual vehicle testing, verification, and calibration testing cycles, reducing development and testing costs.

[0075] 3. The present invention supports virtual parameter calibration and provides system optimization suggestions. BRIEF DESCRIPTION OF THE DRAWINGS

[0076] Figure 1 It is a functional block diagram of an embodiment of the present invention.

[0077] Figure 2 is a flow chart of an embodiment of the present invention. DETAILED DESCRIPTION

[0078] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0079] See also Figure 1 The embodiment of the present invention adopts the hardware-in-the-loop (HiL) method for the simulation test of the ELK system, which consists of three parts: the host computer (workstation), the cabinet (including the slave computer), and the ADAS domain controller (ECU) under test, as follows Figure 1 The test object ELK system runs in the ADAS domain controller. The host computer runs the test management software, which is used to call the simulation test scenario and control system model operation. It communicates with the slave computer via Ethernet. The slave computer runs the Carmaker scenario simulation software, vehicle dynamics model, signal transmission model, and driver board to realize the simulation scenario operation, vehicle dynamics response, CAN communication with the ECU under test, and real-time feedback of test data to the host computer. The ECU runs the ELK application software and issues instructions such as corner control.

[0080] See also Figure 2 The simulation test and evaluation method of the vehicle emergency lane keeping system according to an embodiment of the present invention comprises the following steps:

[0081] In the ELK simulation test, key data of the test vehicle (such as ELK function switch signals, vehicle speed, gear position, etc.), key data of the target vehicle obtained by the sensor model in the simulation environment (such as the lane the target vehicle is in, relative distance / speed / acceleration), and lane line data (lane line type, lane line coefficient, etc.) are transmitted to the ADAS domain controller via the CAN network. The ELK application software integrated in the controller comprehensively determines whether to intervene and the output angle control value when intervening based on the above data. The control value is fed back to the vehicle dynamics software and Carmaker scenario software to achieve closed-loop control of the test vehicle dynamics.

[0082] After completing the HiL environment construction, the ELK test scenario should be established in the simulation software according to the ELK function definition document and test requirement document. Figure 2 The following is a flowchart of a typical ELK simulation test scenario. First, the vehicle power-up operation is simulated. After power-up, the ELK status bits are monitored to ensure they meet the functional definition (e.g., waiting state). The ELK function is then modeled to enable operation and the ELK state is monitored to ensure it transitions correctly (e.g., active state). During this phase, the driver model in the simulation software controls the vehicle's lateral and longitudinal control (steering, braking, and driving). The scenario then sets the driver model to control the vehicle to begin lane departure (steering angle input), causing the test vehicle to begin deviating from the centerline of the original lane. The ELK is monitored for an alarm. If an alarm is issued, the distance from the lane line and the TLC value are output. If the ELK does not intervene during the deviation process, the simulation is terminated when the vehicle completely departs from its original lane, and test data is recorded. If the ELK issues a steering intervention request during the deviation process, control of the test vehicle is immediately switched from the driver model to the ADAS controller. The vehicle dynamics model responds to the steering angle intervention request issued by the ELK, implementing the ELK steering intervention function. During ELK intervention, the ELK steering angle request value is recorded in real time, and the test vehicle's lateral velocity, acceleration, jerk, yaw rate, yaw acceleration, and TLC values ​​are recorded until the ELK intervention is terminated or the lane deviates from the lane.

[0083] A simulation test and evaluation system for ELK in an embodiment of the present invention is used for agile testing and verification throughout the entire function development process. The system evaluates the functions and performance of the ELK system, provides specific scores for each simulation scenario in the ELK system to determine whether system requirements are met, and provides optimization and calibration suggestions to function developers.

[0084] The evaluation of a single scenario is divided into functional indicators and performance indicators. All functional indicators must be achieved, otherwise zero points will be awarded. Performance indicators will be scored based on their achievement.

[0085] The example of ELK system functional indicators is shown in Table 1, which is determined according to the functional definition and test requirements document of the ELK system. The evaluation result of each indicator is f i .

[0086]

[0087]

[0088] 1) f1 is used to monitor whether the ELK function state machine is consistent with the function definition (such as waiting, activation, inhibition, fault state, etc.) during the entire simulation process

[0089] 2) f2 means that when the lane line is dotted, there is an adjacent lane (i.e., lane change is allowed), and there is no collision risk, the ELK function should not interfere with the test vehicle's lane change and should not issue an alarm.

[0090] 3) f3 means that when the lane line is non-dashed (solid line, double solid line, etc.), there is an adjacent lane (i.e., lane change is allowed), the driver turns on the turn signal and there is no collision risk, it should be considered that the driver has the intention to actively change lanes. The ELK function should not interfere with the test vehicle's lane change and should not issue an alarm.

[0091] 4) f4 refers to when the lane line is non-dashed (solid line, double solid line, etc.), there is an adjacent lane (i.e., lane change is allowed), and the turn signal is not on. This is a typical ELK operating condition, and an alarm should be issued and the test vehicle should be intervened to change lanes.

[0092] 5) f5 refers to the situation where the lane line is dotted (solid, double solid, etc.), there is an adjacent lane (i.e., lane change is possible), and there is a collision risk when the test vehicle deviates from the original lane. This is a typical ELK operating condition, and an alarm should be issued and the test vehicle should be intervened to change lanes.

[0093] 6) f6 refers to when the lane line is dotted (solid, double solid, etc.), there is an adjacent lane (i.e., lane change is allowed), and there is a collision risk when the test vehicle deviates from the original lane. This is a typical ELK operating condition, and an alarm should be issued and the test vehicle should be intervened to change lanes.

[0094] 7) f7 means the test vehicle deviates from the curb. This is a typical ELK operating condition and an alarm should be issued and the test vehicle should be intervened to change lanes.

[0095] 8) f8 is the basic requirement of ELK function. During ELK intervention, the test vehicle must not deviate from the original lane and the crossing line should be less than 30cm.

[0096] 9) f9 Based on the ELK function definition and driving safety requirements, constrain the heading angle of the test vehicle after the ELK intervention is completed, so that the test vehicle does not deviate from the original lane within 50m after the ELK intervention is completed

[0097] The functional achievement score F (Function) is introduced to evaluate the functional indicators of a simulation scenario test of ELK, as shown below (1):

[0098]

[0099] If and only if f i When (i=1, 2, 3, 4, ...) are all 1 and all functional indicators of ELK are met, F=1.

[0100] The performance indicators of the ELK system are shown in Table 2:

[0101] Performance evaluation indicators symbol TLC during alarm times <![CDATA[TLC w a rning ]]> Distance between alarm time and lane line <![CDATA[D warning ]]> Turn to the TLC of the intervention moment <![CDATA[TLC steering ]]> Steering intervention time and lane line distance <![CDATA[D steering ]]> lateral speed <![CDATA[v y ]]> lateral acceleration <![CDATA[a y ]]> Lateral acceleration rate of change <![CDATA[a″ y ]]> Yaw angular velocity <![CDATA[Yaw rate ]]> Yaw rate of change <![CDATA[Yaw acc ]]> The closest distance between the test vehicle and the lane line <![CDATA[LatDisplace min ]]> Yaw angle after ELK is enabled <![CDATA[Yaw terminate ]]> … …

[0102] 1) TLC warnig Used to calculate the timing of the test vehicle deviation alarm issued by the ELK system, and to evaluate whether the alarm timing is too early or too late

[0103] 2)D warning Used to evaluate the time when the test vehicle deviates from the lane line and the distance between the lane line and the time when the ELK system issues the warning, and to evaluate whether the warning distance is too early or too late.

[0104] 3) TLC steering Used to calculate the timing of the ELK system's steering intervention and to evaluate whether the steering intervention is too early or too late

[0105] 4)D steering Used to evaluate the distance between the test vehicle and the lane line when the ELK system intervenes in steering, and to evaluate whether the steering intervention distance is too early or too late

[0106] 5)v y Refers to the test vehicle simulating lateral speed and evaluating its absolute value

[0107] 6)a y Refers to the lateral acceleration of the test vehicle, which should not be greater than 6m / s^2 (reference value), otherwise it will have a significant impact on driver comfort

[0108] 7)a″ y Refers to the rate of change of the lateral acceleration of the test vehicle, which should not usually exceed 8m / s^3, otherwise it will have a significant impact on the driver's comfort

[0109] 8) Yaw rate Refers to the yaw rate of the test vehicle during simulation

[0110] 9) Yaw acc Refers to the yaw rate of change during the test vehicle simulation process

[0111] 10)LatDisplace min This value is used to calculate the effect of ELK intervention on the lateral movement of the vehicle. In principle, no negative value should appear, i.e. the vehicle crosses the line, and a safe lateral distance of 10 cm should be left as much as possible.

[0112] 11) Yaw terminate Used to calculate the yaw angle of the test vehicle after ELK intervention is completed (relative to the original lane centerline)

[0113] Based on the test purpose and test content of each simulation scenario, the data of each performance indicator is evaluated, with a full score of 10 points. The factors affecting the scoring results are mainly determined by the following aspects:

[0114] 1) Driving safety mainly refers to the difference in scene complexity and risk. The timing of ELK intervention should be appropriate and not too late to avoid the risk of vehicles crossing the line.

[0115] 2) Driving comfort, which mainly refers to the requirement that the maximum values ​​of vehicle acceleration and acceleration change rate are within a reasonable range while being able to avoid collision risks.

[0116] 3) User complaint level: This mainly refers to the fact that the ELK system will only intervene when there is a certain risk, and the intervention should not be too early to avoid frequent intervention and user complaints.

[0117] Based on the first performance evaluation index TLC warning For example, the crossing time is equal to the distance between the outer edge of the vehicle's front wheel and the lane line divided by the vehicle's lateral speed. The expected value of this indicator is related to the road conditions and the vehicle's lateral speed. The expected value is small on a straight road, large when the vehicle deviates to the outside of the curve, and moderate when it deviates to the inside of the curve; the greater the lateral speed, the greater the expected value, which means an early alarm. Table 3 below shows the TLC when a vehicle deviates from a straight road at a lateral speed of 1m / s. warning The evaluation criteria for TLC warning The score is p1.

[0118]

[0119] Similarly, the other 10 performance evaluation indicators are scored to obtain the specific score p of each performance evaluation indicator in each simulation scenario. i , as shown in Table 4 below.

[0120] Performance evaluation indicators symbol Score TLC during alarm times <![CDATA[TLC warning ]]> <![CDATA[p1]]> Distance between alarm time and lane line <![CDATA[D warning ]]> <![CDATA[p2]]> Turn to the TLC of the intervention moment <![CDATA[TLC steering ]]> <![CDATA[p3]]> Steering intervention time and lane line distance <![CDATA[D steering ]]> <![CDATA[p4]]> lateral speed <![CDATA[v y ]]> <![CDATA[p5]]> lateral acceleration <![CDATA[a y ]]> <![CDATA[p6]]> Lateral acceleration rate of change <![CDATA[d y ]]> <![CDATA[p7]]> Yaw angular velocity <![CDATA[Yaw rate ]]> <![CDATA[p8]]> Yaw rate of change <![CDATA[Yaw acc ]]> <![CDATA[p9]]> Maximum lateral deviation distance of the test vehicle <![CDATA[LatDisplace max ]]> <![CDATA[p 10 ]]> Yaw angle after ELK is enabled <![CDATA[Yaw terminate ]]> <![CDATA[p 11 ]]>

[0121] The following is a calculation of the weights of each performance indicator score to obtain the performance evaluation score of the ELK system in a certain simulation scenario. Based on the dimensions of driving safety, driving comfort, and user complaints, the performance indicator weights of the ELK system are comprehensively calculated. The following are the first and second most important performance evaluation indicators of the ELK system.

[0122] 1) TLC steering It is an important indicator for judging whether the system intervention time is appropriate, and it is usually recommended to give it the highest weight value (18%).

[0123] 2)LatDisplace min It is an important indicator for judging the effectiveness of the system's steering intervention. It is used to calculate the minimum lateral distance between the vehicle and the lane line and to determine whether the vehicle has crossed the line. It is usually recommended to give it the second highest weight (15%).

[0124] This yields the weight values ​​corresponding to each performance evaluation parameter:

[0125] ω i (∑ω i =1,ω i >0; i=1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11,...),

[0126] The performance evaluation of ELK in a certain scenario is P (Performance):

[0127]

[0128] The comprehensive evaluation of ELK in a certain scenario is S (Score):

[0129] S=F*P

[0130] S is the comprehensive evaluation score, which requires that all functional indicators are met, and the specific score is determined by the performance evaluation indicators.

[0131] The above embodiments are intended only to illustrate the design concepts and features of the present invention. Their purpose is to enable those skilled in the art to understand the contents of the present invention and implement them accordingly. The scope of protection of the present invention is not limited to the above embodiments. Therefore, any equivalent changes or modifications made based on the principles and design concepts disclosed in the present invention are within the scope of protection of the present invention.

Claims

1. A simulation test and evaluation method for a vehicle emergency lane keeping system, characterized by: The following steps are involved: S0: Build a simulation test and evaluation system for a hardware-in-the-loop (HiL) environment, including a host computer, a slave computer, and the ADAS domain controller under test. The host computer runs test management software, and the host and slave computers communicate via Ethernet. The slave computers run Carmaker scenario simulation software, vehicle dynamics models, signal transmission models, and driver boards to implement simulation scenario operation, vehicle dynamics response, CAN communication with the ECU under test, and real-time test data feedback to the host computer. The ADAS domain controller runs the test object ELK. S1: Create an ELK test scenario in the simulation software based on the ELK function definition document and test requirement document. The specific steps are as follows: S11: Simulate the vehicle power-on operation, and monitor the ELK status bit to see if it meets the function definition after power-on is completed; S12: The model ELK function is turned on and the ELK status is monitored to see if it is correctly switched. During this stage, the driver model in the simulation software controls the vehicle in both the lateral and longitudinal directions. This control includes steering, braking, and driving. S13: The driver model is set in the scene to input a steering angle to the control vehicle, and the test vehicle begins to deviate from the center line of the original lane; Determine whether the monitoring ELK issues an alarm. If an alarm is issued, output the distance to the lane line and the TLC value at the time of the alarm; If ELK does not intervene during the deviation process, the simulation will be stopped when the vehicle completely deviates from the original lane and the test data will be recorded; If the ELK issues a steering intervention request during the deviation process, the test vehicle is immediately switched from driver model control to ADAS controller control, and the vehicle dynamics model responds to the angle intervention request value issued by the ELK to implement the ELK steering intervention function; During ELK intervention, the ELK's requested turning angle is recorded in real time, as well as the test vehicle's lateral velocity, acceleration, jerk, yaw rate, yaw acceleration, and TLC values, until the ELK intervention is terminated or the vehicle deviates from its lane. S2: Evaluate the functions and performance of the ELK system and give specific scores for each simulation scenario in the ELK system to determine whether it meets the system requirements.

2. The simulation test and evaluation method for a vehicle emergency lane keeping system according to claim 1, characterized in that: In the step S0, the specific steps are: In the ELK simulation test, the key data of the test vehicle, the key data of the target vehicle obtained by the sensor model in the simulation environment, and the lane line data are transmitted to the ADAS domain controller via the CAN network; The controller's integrated ELK application software uses this data to comprehensively determine whether to intervene and outputs the steering angle control value when intervening. The control value is then fed back to the vehicle dynamics software and Carmaker scenario software to achieve closed-loop control of the test vehicle's dynamics. Key data of the test vehicle include ELK function switch signal, vehicle speed and gear position; The key data of the target vehicle obtained by the sensor model in the simulation environment include the lane where the target vehicle is located, relative distance, relative speed and relative acceleration; Lane line data includes lane line type and lane line coefficient.

3. The simulation test and evaluation method for a vehicle emergency lane keeping system according to claim 1, characterized in that: In step S2, the single scenario evaluation includes functional indicators and performance indicators; if all functional indicators are met, 1 point is recorded, and if not met, 0 point is recorded; the performance indicators are scored according to the degree of achievement; Assume that the evaluation results of the functional indicators of the ELK system are , specifically: Used to monitor whether the ELK function state machine is consistent with the function definition during the entire simulation process, including waiting, activation, inhibition and fault states; This means that when the lane line is dotted, there is an adjacent lane, and there is no collision risk, the ELK function will not interfere with the test vehicle's lane change and will not issue an alarm; When the lane marking is not a dotted line, there is an adjacent lane, the driver has turned on the turn signal, and there is no collision risk, the ELK function is considered to have the intention to change lanes. It does not interfere with the test vehicle's lane change and does not issue an alarm. When the lane line is not a dotted line, there is an adjacent lane, and the turn signal is not on, it is a typical ELK operating condition, an alarm is issued and the test vehicle is intervened to change lanes; When the lane line is dotted, there is an adjacent lane, and the test vehicle deviates from the original lane, there is a collision risk. This is a typical ELK working condition, and an alarm is issued and the test vehicle is intervened to change lanes; When the lane line is dotted, there is an adjacent lane, and the test vehicle deviates from the original lane, there is a collision risk. This is a typical ELK working condition, and an alarm is issued and the test vehicle is intervened to change lanes; The test vehicle deviates from the curb, which is a typical ELK operating condition. An alarm is issued and the test vehicle is intervened to change lanes. This is the basic requirement of the ELK function. During the ELK intervention process, the test vehicle must not deviate from the original lane and the crossing line should be less than 30cm; Based on the ELK function definition and driving safety requirements, the heading angle of the test vehicle is constrained after the ELK intervention is completed, so that the test vehicle does not deviate from the original lane within 50m after the ELK intervention is completed.

4. The simulation test and evaluation method for a vehicle emergency lane keeping system according to claim 3, characterized in that: In step S2, the function achievement score F is introduced to evaluate the function indicators of a certain simulation scenario test of ELK: If and only if When (i=1,2,3,4,…) are all 1 and all functional indicators of ELK are met, ; The specific performance indicators of the ELK system are: Used to calculate the timing of the test vehicle deviation alarm issued by the ELK system and to evaluate whether the alarm timing is too early or too late; Used to evaluate the time when the test vehicle's departure warning is issued by the ELK system and the distance from the lane line, and to evaluate whether the warning distance is too early or too late; Used to calculate the timing of the ELK system's steering intervention and to evaluate whether the steering intervention is too early or too late; Used to assess the distance between the test vehicle and the lane line when the ELK system intervenes in steering, and to evaluate whether the steering intervention distance is too early or too late; Refers to the simulated lateral speed of the test vehicle and the assessment of its absolute value; Refers to the lateral acceleration of the test vehicle, not exceeding 6m / s^2; Refers to the rate of change of the lateral acceleration of the test vehicle, which shall not exceed 8m / s^3; Refers to the yaw angular velocity of the test vehicle during simulation; Refers to the rate of change of yaw angular velocity during the simulation of the test vehicle; Used to calculate the effect of ELK intervention on the vehicle's lateral movement. A non-negative value means the vehicle must not cross the line and must leave a safe lateral distance of 10 cm. Used to calculate the yaw angle of the test vehicle relative to the center line of the original lane after ELK intervention is completed.

5. The simulation test and evaluation method for a vehicle emergency lane keeping system according to claim 4, characterized in that: In step S2, the data of each performance indicator is evaluated based on the test purpose and test content of each simulation scenario, with a full score of 10 points; Factors affecting the rating results include driving safety, driving comfort and user complaints; Driving safety means that ELK intervention should not be too late, based on the complexity and risk of the scene, to avoid the risk of vehicles crossing the line; Driving comfort means that the maximum values ​​of vehicle acceleration and acceleration change rate must be within a reasonable range while avoiding collision risks. User complaint level means that when there is a certain risk, the ELK system should not intervene too early to avoid frequent intervention and steering that may cause user complaints.

6. The simulation test and evaluation method for a vehicle emergency lane keeping system according to claim 5, characterized in that: In step S2, the weights of the performance index scores are calculated to obtain the performance evaluation score of the ELK system in a certain simulation scenario. The performance index weights of the ELK system are comprehensively obtained based on the dimensions including driving safety, driving comfort, and user complaints: It is an important indicator for judging whether the system's intervention moment is appropriate and is given the highest weight; It is an important indicator for judging the effectiveness of the system's steering intervention. It is used to calculate the minimum lateral distance between the vehicle and the lane line and to determine whether the vehicle has crossed the lane line. It is given the second highest weight. This yields the weight values ​​corresponding to each performance evaluation parameter: （ ， >0；i=1,2,3,4,5,6,7,8,9,10,11,…）， The performance evaluation of ELK in a certain scenario is : The comprehensive evaluation of ELK in a certain scenario is : S is the comprehensive evaluation score, which requires that all functional indicators are met, and the score is determined by the performance evaluation indicators.

7. A simulation test and evaluation system for the simulation test and evaluation method of a vehicle emergency lane keeping system according to any one of claims 1 to 6, characterized in that: Adopting the hardware-in-the-loop (HiL) approach, including the host computer, the slave computer, and the ADAS domain controller to be tested; The host computer runs the test management software, which is used to call the simulation test scenarios and control the operation of the system model. The host computer and the slave computer communicate through Ethernet; The lower computer runs the Carmaker scenario simulation software, vehicle dynamics model, signal transmission model, and driver board to implement simulation scenario operation, vehicle dynamics response, CAN communication with the ECU under test, and feed back test data to the upper computer in real time. The ADAS domain controller runs the test object ELK to issue control instructions.

8. A computer storage medium, characterized in that: A computer program executable by a computer processor is stored therein, and the computer program executes the simulation test and evaluation method for the vehicle emergency lane keeping system according to any one of claims 1 to 6.

Images