Vehicle limit performance automation test method and device, and electronic equipment

Abstract

The present disclosure provides a vehicle limit performance automation test method and device and electronic equipment, and relates to the technical field of vehicle testing. Through autonomous perception of the vehicle, the road dynamic adhesion coefficient and the test geometric boundary are calculated, the dynamic adhesion coefficient is used to construct a dynamic constraint, and a physically feasible reference trajectory and an initial speed curve are generated through optimal control. The trajectory is tested and closed-loop iteration is automatically converged to the limit passing speed. A normalized objective evaluation score is generated based on multi-dimensional core data. Therefore, the problems of insufficient limit exploration, strong subjectivity of the evaluation system and lack of unified objective standard due to reliance on artificial preset parameters and subjective driving of professional drivers in the prior art can be solved. The technical effects of full autonomy without artificial intervention, accurate approximation of vehicle physical limits, objective and consistent results, and quantifiable results, and horizontally comparable test results under different road conditions are achieved.

Description

Technical Field

[0001] This disclosure relates to the field of vehicle testing technology, and in particular to an automated method and apparatus for testing the extreme performance of vehicles, as well as electronic equipment. Background Technology

[0002] Vehicle handling stability testing, a core component of automotive R&D, is widely used in moose tests and slalom assessments. With the development of intelligent connected vehicle technology, a dynamic performance testing system has been constructed through the collaborative operation of onboard sensors, trajectory planning, and chassis control. Specifically, this system covers the entire process from environmental perception and path generation to extreme obstacle avoidance, including key aspects such as cone recognition, speed curve planning, and stability assessment, aiming to quantify a vehicle's emergency obstacle avoidance capabilities.

[0003] However, existing testing methods rely directly on the subjective driving and experience-based preset parameters of professional drivers, without establishing an autonomous perception and iterative mechanism based on physical constraints. Due to differences in driver psychological qualities and limitations imposed by fear, tests struggle to truly approach the vehicle's physical limits, especially on low-friction surfaces. This results in poor consistency of results, insufficient exploration of limits, and a lack of unified objective evaluation standards, severely impacting the quantitative comparison of chassis tuning across different vehicle models and the efficiency of electronic stability system calibration. Summary of the Invention

[0004] This disclosure provides an automated testing method, apparatus, and electronic device for vehicle extreme performance. Its main purpose is to at least partially solve one of the technical problems in related technologies.

[0005] According to a first aspect of this disclosure, an automated method for testing the extreme performance of a vehicle is provided, comprising:

[0006] The vehicle autonomously excites and collects vehicle response data through onboard sensors to calculate the dynamic adhesion coefficient of the current road surface. At the same time, it autonomously identifies the geometric boundaries of the test area and generates a sequence of necessary points for trajectory planning. A vehicle dynamics constraint model is constructed based on the dynamic adhesion coefficient. Using geometric boundaries and a sequence of necessary points, a reference trajectory that satisfies physical feasibility and an initial velocity curve corresponding to the reference trajectory are generated through optimal control solution. The vehicle is controlled to perform the test according to the reference trajectory and initial speed curve. Based on the comparison between the real-time monitored vehicle status response and the preset limit criterion, the target speed is automatically adjusted and multiple rounds of iterative exploration are carried out until it converges to the limit passing speed. Based on the converged limit speed, dynamic adhesion coefficient, and trajectory tracking accuracy during the test, a normalized objective evaluation score is calculated and generated.

[0007] According to a second aspect of this disclosure, an automated testing apparatus for vehicle extreme performance is provided, comprising: The calculation unit is used to autonomously excite and collect vehicle response data through onboard sensors to calculate the dynamic adhesion coefficient of the current road surface, and at the same time autonomously identify the geometric boundary of the test area to generate a sequence of necessary points for trajectory planning. The generation unit is used to construct a vehicle dynamics constraint model based on the dynamic adhesion coefficient, and to generate a reference trajectory that meets physical feasibility and the corresponding initial velocity curve through optimal control by using geometric boundaries and a sequence of necessary points. The test unit is used to control the vehicle to perform tests according to the reference trajectory and initial speed curve. Based on the comparison results of the real-time monitored vehicle state response and the preset limit criteria, it automatically adjusts the target speed and conducts multiple rounds of iterative exploration until it converges to the limit passing speed. The calculation unit is used to calculate and generate a normalized objective evaluation score based on the converged limit pass speed, dynamic adhesion coefficient, and trajectory tracking accuracy during the test process.

[0008] According to a third aspect of this disclosure, an electronic device is provided, comprising: At least one processor; and A memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor to enable the at least one processor to perform the method described in the first aspect above.

[0009] According to a fourth aspect of this disclosure, a non-transitory computer-readable storage medium is provided storing computer instructions, wherein the computer instructions are configured to cause the computer to perform the method described in the first aspect above.

[0010] According to a fifth aspect of this disclosure, a computer program product is provided, comprising a computer program that, when executed by a processor, implements the method described in the first aspect above.

[0011] The automated testing method, apparatus, and electronic equipment for vehicle limit performance disclosed herein enable the vehicle to autonomously perceive and calculate the dynamic adhesion coefficient of the road surface and the test geometric boundary, construct dynamic constraints based on the dynamic adhesion coefficient, generate a physically feasible reference trajectory and initial velocity curve through optimal control, test along the trajectory and automatically converge to the limit passing speed through closed-loop iteration, and generate a normalized objective evaluation score based on multi-dimensional core data. Therefore, it can solve the problems of existing technologies that rely on manually preset parameters and subjective driving by professional drivers, insufficient limit exploration, strong subjectivity of the evaluation system, and lack of unified objective standards. It achieves the technical effects of fully autonomous testing without human intervention, accurately approximating the physical limits of the vehicle, objective, consistent, and quantifiable results, and horizontal comparability of test results under different road conditions.

[0012] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of this disclosure, nor is it intended to limit the scope of this disclosure. Other features of this disclosure will become readily apparent from the following description. Attached Figure Description

[0013] The accompanying drawings are provided to better understand this solution and do not constitute a limitation of this disclosure. Wherein: Figure 1 A flowchart illustrating an automated testing method for vehicle extreme performance provided in this embodiment of the present disclosure; Figure 2 This is a schematic diagram of the structure of an automated vehicle extreme performance testing device provided in an embodiment of the present disclosure; Figure 3 A schematic block diagram of an example electronic device provided for embodiments of this disclosure. Detailed Implementation

[0014] The exemplary embodiments of this disclosure are described below with reference to the accompanying drawings, including various details of the embodiments to aid understanding, and should be considered merely exemplary. Therefore, those skilled in the art will recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of this disclosure. Similarly, for clarity and brevity, descriptions of well-known functions and structures are omitted in the following description.

[0015] The embodiments disclosed herein, at every technical stage of the data lifecycle, including but not limited to data collection, transmission, storage, computation, use, disclosure, and destruction, are fundamentally based on strict adherence to and embedding of current laws, regulations, and regulatory requirements in their system architecture, protocols, and process controls. At the design level, the solution ensures, through systematic rules and strategies, that all processing activities automatically adhere to the principles of legality, legitimacy, necessity, and good faith, and technically implements core rules such as clear purpose, minimum necessity, transparency, and security.

[0016] For any data collection, processing, or other activities involved in the embodiments of this disclosure, corresponding verification, tracking, and constraint mechanisms are implemented at the system level to ensure that their execution has a clear legal basis or contractual foundation, and to automatically trigger and record the corresponding notification process. The processing purpose of related data is bound to its specific use at the metadata layer, and is strictly limited through the system's embedded flow strategy and access control model, thereby ensuring that data is accessed and used only within the scope necessary to achieve the initial collection purpose and as determined by technical criteria. The system has a multi-layered authorization management and compliance audit mechanism to ensure that related data will not be used for any other purpose without separate legal permission or valid separate consent from the information subject. This solution natively supports and protects the information subject's various legal rights to their data in its technical implementation, and provides standardized interfaces and automated processes to achieve efficient exercise of these rights.

[0017] The following description, with reference to the accompanying drawings, outlines an automated method, apparatus, and electronic device for testing the extreme performance of vehicles according to embodiments of this disclosure.

[0018] Figure 1 This is a flowchart illustrating an automated testing method for vehicle limit performance provided in an embodiment of this disclosure.

[0019] like Figure 1 As shown, the method includes the following steps: Step 101: The vehicle autonomously excites and collects vehicle response data through onboard sensors to calculate the dynamic adhesion coefficient of the current road surface. At the same time, it autonomously identifies the geometric boundary of the test area and generates a sequence of necessary points for trajectory planning.

[0020] In the embodiments of this disclosure, the chassis execution module first applies autonomous excitation to the vehicle, and simultaneously collects real-time response data of the vehicle under this excitation through onboard sensors. Then, the excitation signal and response data are input into a preset identification model to calculate the dynamic adhesion coefficient of the current road surface. The entire process does not require manual input of adhesion parameters. Simultaneously, the onboard sensors continuously collect raw environmental data of the test area. The environmental perception module performs target recognition and boundary extraction on the data, autonomously determines the geometric boundary of the test area, and then, based on the path requirements of the test conditions, filters and generates a sequence of necessary points for adaptive trajectory planning from within the geometric boundary range. As an example, autonomous excitation can use small-amplitude steering excitation or micro-pulse braking excitation. The vehicle response data includes lateral acceleration, longitudinal deceleration, wheel speed difference, etc.; the onboard sensors can be cameras or lidar; the geometric boundary includes the width of the test channel, the position of the cones, the coordinates of the entrance and exit of the test area, etc.; and the sequence of necessary points is the core path control point for the vehicle to pass through the test area.

[0021] By coordinating autonomous excitation calculation and autonomous environmental identification, the road surface adhesion coefficient and test geometric boundary can be obtained without human intervention, ensuring the real-time nature and objectivity of the test data and providing accurate and reliable input for subsequent trajectory planning.

[0022] Step 102: Construct a vehicle dynamics constraint model based on the dynamic adhesion coefficient, and use the geometric boundary and the sequence of necessary points to generate a reference trajectory that meets physical feasibility and the initial velocity curve corresponding to the reference trajectory through optimal control solution.

[0023] In the embodiments of this disclosure, a vehicle dynamics constraint model corresponding to the current road surface conditions is constructed using the dynamic adhesion coefficient as the core parameter, clarifying the grip force boundary and kinematic constraints during vehicle movement. Subsequently, the processor uses the geometric boundary as the spatial constraint for path planning and the sequence of necessary points as the core control points for path fitting. Based on the above constraints and control points, an optimal control solution model is constructed, with trajectory smoothness, driving stability, and physical feasibility as optimization objectives. The model is solved using a preset optimal control algorithm to generate a reference trajectory with continuous position and curvature that perfectly matches the vehicle dynamics constraints. Simultaneously, the processor combines the curvature distribution characteristics of the reference trajectory with the vehicle dynamics constraint model to simultaneously solve for an initial velocity curve that matches the reference trajectory one-to-one and conforms to the physical rules of vehicle movement. As an example, the vehicle dynamics constraint model may include friction ellipse constraints, velocity-curvature constraints, etc. The optimal control algorithm may employ polynomial spline optimization or dynamic programming algorithms. The initial velocity curve is adaptively adjusted with the trajectory curvature, with the velocity decreasing in the curve segment and increasing in the straight segment.

[0024] By constructing dynamic constraints using dynamic adhesion coefficients and combining them with optimal control solutions, the reference trajectory and initial velocity curve are precisely adapted to the actual adhesion characteristics of the road surface. This fundamentally ensures the physical feasibility of the planning results and provides a stable and executable driving benchmark for automated vehicle testing.

[0025] Step 103: Control the vehicle to perform the test according to the reference trajectory and initial speed curve. Based on the comparison results of the real-time monitored vehicle status response and the preset limit criteria, automatically adjust the target speed and conduct multiple rounds of iterative exploration until convergence to the limit passing speed.

[0026] In the embodiments of this disclosure, trajectory tracking and speed control commands are issued to the vehicle chassis execution module to drive the vehicle to strictly follow the reference trajectory and initial speed curve to carry out the test operation. During the test, the on-board status monitoring unit collects the vehicle's driving status response data in real time. The closed-loop control processor compares this real-time status data with preset limit criteria item by item to quickly determine whether the current driving state has reached the vehicle's limit. Based on the comparison results, the processor automatically executes the target speed adjustment operation. If the limit criteria are not triggered, the target speed is increased; if the limit criteria are triggered, the target speed is decreased, and the vehicle is controlled to repeat the test at the adjusted target speed. The processor continuously carries out the above-mentioned monitoring, comparison, adjustment, and retesting multi-round iterative process, continuously narrowing the target speed value range until the target speed tends to stabilize, and finally converges to obtain the limit passing speed under the current test conditions. As an example, the vehicle status response data includes yaw rate, center of gravity sideslip angle, wheel speed difference, trajectory tracking offset, etc., and the preset limit criteria include sideslip angle threshold, trajectory offset threshold, cone contact judgment condition, etc. Iterative adjustment can be implemented using the bisection method or the golden section method.

[0027] Through a closed-loop iterative autonomous testing and control method, the vehicle can accurately converge to its limit speed without human intervention, effectively improving the repeatability and accuracy of limit performance testing and avoiding interference from human factors on the test results.

[0028] Step 104: Based on the converged limit speed, dynamic adhesion coefficient, and trajectory tracking accuracy during the test, calculate and generate a normalized objective evaluation score.

[0029] In the embodiments of this disclosure, the core calculation inputs are the ultimate passing speed obtained in step 103, the dynamic adhesion coefficient calculated in step 101, and the trajectory tracking accuracy collected and statistically analyzed throughout the test. In step 104, the on-board evaluation calculation processor calls a preset normalized evaluation calculation model, substitutes the ultimate passing speed, dynamic adhesion coefficient, and trajectory tracking accuracy into the model to complete the numerical calculation, and eliminates the evaluation bias caused by differences in road surface adhesion conditions and test conditions through dimensionless processing, directly generating a normalized objective evaluation score that can be used for horizontal comparison. As an example, the normalized evaluation calculation model can adopt a linear weighted or ratio calculation form, the trajectory tracking accuracy is used to correct the final score, and the generated evaluation score has a fixed value range, which can realize a unified quantitative evaluation of the vehicle's ultimate performance under different road surfaces and different test scenarios.

[0030] Normalized objective evaluation scores are generated based on multi-dimensional measured data, completely eliminating subjective evaluation interference. This allows the results of different road surfaces and different test conditions to have a unified comparison benchmark, providing an objective and quantitative evaluation basis for vehicle limit performance assessment.

[0031] In the embodiments involved in this application, there are various feasible specific implementation methods. To clearly and completely illustrate the technical solutions of this disclosure, the implementation methods listed below are merely exemplary and do not constitute a limitation on the scope of protection of this disclosure. That is, in addition to the implementation methods described below, other implementation methods that can be obtained by those skilled in the art based on the technical content disclosed in this disclosure through reasonable logical analysis, reasoning, or limited experimentation should also be covered within the scope of protection of this disclosure. The following specifically describes some exemplary implementation methods: As a specific implementation of this disclosure, based on the basic scheme, the vehicle-mounted sensors autonomously excite and collect vehicle response data to calculate the dynamic adhesion coefficient of the current road surface. At the same time, the system autonomously identifies the geometric boundaries of the test area and generates a sequence of necessary points for trajectory planning. This is further defined as follows: the steering system autonomously applies a sinusoidal sweep frequency input or the braking system applies a micro-braking pulse as an excitation signal, and the vehicle's lateral acceleration or longitudinal deceleration is collected as vehicle response data. The instantaneous peak adhesion coefficient is calculated in real time by comparing the excitation signal and the vehicle response data based on the recursive least squares method. The vehicle-mounted vision or point cloud sensors identify the characteristics of the cones to establish a cone point cloud, automatically distinguish different functional areas or identify the cone arrangement axis, and calculate the cone spacing, lane width and entrance / exit location information in real time, thereby generating a sequence of necessary points that the vehicle must pass through.

[0032] Specifically, in one optional refined implementation, the specific implementation methods for autonomous excitation, vehicle response acquisition, adhesion coefficient calculation, and necessary point sequence generation are further defined. This implementation abandons the generalized excitation and perception logic, and instead, the vehicle chassis execution unit outputs excitation signals in a directional manner, simultaneously acquiring matching dynamic response data, and completing the road adhesion coefficient calculation through a dedicated parameter identification algorithm; at the same time, the environmental perception unit directionally identifies the test cone target, analyzes the geometric features of the test area, and generates standardized path control points.

[0033] In some specific embodiments, the excitation signal is autonomously applied by the steering system with a sinusoidal sweep frequency of 1Hz and an amplitude of 3°-5°, or by the braking system applying micro-brake pulses; the vehicle response data acquisition module simultaneously collects the vehicle's lateral acceleration and longitudinal deceleration as core response quantities. Based on the recursive least squares method, the system compares and fits the excitation signal and vehicle response data in real time to directly calculate the instantaneous peak adhesion coefficient μ of the current road surface and tire matching. At the environmental perception level, the system identifies the color, shape, and position characteristics of the cones using onboard visual sensors or lidar point cloud sensors, establishing a complete cone point cloud; for the moose test condition, it automatically distinguishes the entrance, expansion, and exit areas; for the cone-around test condition, it identifies the cone arrangement axis and calculates geometric parameters such as cone spacing, lane width, and entrance / exit positions in real time, generating a sequence of necessary points for vehicle passage based on these parameters. In addition to the above excitation methods and sensing devices, similar alternative methods such as small-amplitude suspension excitation and millimeter-wave radar sensing can also be used to achieve the same function.

[0034] By combining directional excitation with recursive least squares method, the real-time performance and accuracy of road surface adhesion coefficient calculation are greatly improved. At the same time, the parameters of piles and functional areas are accurately identified to ensure that the sequence of necessary points fully matches the geometric constraints of the test conditions.

[0035] As a specific implementation of this disclosure, based on the basic scheme, a vehicle dynamics constraint model is constructed based on the dynamic adhesion coefficient. Using geometric boundaries and a sequence of necessary points, an optimal control solution is used to generate a reference trajectory that satisfies physical feasibility and an initial velocity curve corresponding to the reference trajectory. Further, the following is defined: a friction ellipse model is constructed based on the instantaneous peak adhesion coefficient, and the maximum permissible lateral acceleration and velocity-curvature constraint relationship for any path point under a given path curvature are derived; the necessary point sequence is adaptively shrunk according to the instantaneous peak adhesion coefficient. When the instantaneous peak adhesion coefficient is lower than a preset threshold, the necessary points are shrunk towards the centerline of the channel; a fifth-order polynomial spline curve is used to fit the adaptively shrunk necessary point sequence, with the optimization objective being to minimize the distance error between the path point and the necessary point, as well as the integral of the rate of curvature change along the path, generating a reference path with continuous position, tangential angle, and curvature; based on the coupling constraints of longitudinal acceleration and lateral acceleration defined by the friction ellipse model, a dynamic programming algorithm is used to solve for the optimal velocity curve that minimizes the total time to traverse the entire path, which serves as the initial velocity curve.

[0036] Specifically, in one optional refined implementation, the construction of dynamic constraints, adaptive processing of necessary points, fitting of reference trajectory, and solution of initial velocity curve are further specified. This implementation uses the instantaneous peak adhesion coefficient as the core input and deeply integrates dynamic constraints, geometric adaptive adjustment, and optimal control algorithms to ensure that the reference trajectory and initial velocity curve perfectly match the vehicle's physical driving limits.

[0037] In some specific embodiments, a friction ellipse model is constructed based on the instantaneous peak adhesion coefficient μ, and the tire longitudinal force... With lateral force satisfy:

[0038] Among them, the lateral force limit , The vertical load on the tires is estimated from the vehicle's condition; based on this, the maximum permissible lateral acceleration is derived:

[0039] in This is the load transfer correction factor, which is calculated in real time by the system based on the vehicle's center of gravity height, track width, and suspension characteristics.

[0040] At any path point, if the path curvature is Then the maximum permissible speed satisfies:

[0041] The system adaptively shrinks the sequence of necessary points based on the μ value. When μ is below a preset threshold, the necessary points are shrunk towards the centerline of the channel, reserving a safety margin for lateral slippage. A fifth-order polynomial spline curve is used to fit the shrunken sequence of necessary points, generating a reference path with continuous position, tangential angle, and curvature as the objective function. At the velocity planning level, based on the coupling constraints of longitudinal and lateral acceleration of the friction ellipse, a dynamic programming algorithm is used to solve for the optimal velocity curve with the shortest total travel time, which serves as the initial velocity curve. Besides dynamic programming, equivalent algorithms such as model predictive control can also be used to solve for the velocity.

[0042] By accurately binding the road surface adhesion characteristics through the friction ellipse model, the adaptive inward contraction avoids the risk of low-adhesion sideslip, the fifth-order polynomial spline ensures the smoothness of the trajectory, and dynamic programming makes the speed curve conform to the dynamic constraints and optimize the traffic efficiency.

[0043] As a specific embodiment of this disclosure, based on the basic scheme, it is further defined as follows: the optimization objective is to minimize the distance error between path points and necessary points, as well as the integral of the rate of curvature change along the path. The optimization objective function is:

[0044] in, For path points, As a necessary point, For curvature, Let the arc length be , and These are the weighting coefficients.

[0045] Specifically, regarding the reference path generation optimization objective in step S102 (the self-generation step of limit trajectory and velocity) in the main embodiment, in an optional refined implementation, the optimization objective of the fifth-order polynomial spline fitting is explicitly quantified and limited, and the path fitting accuracy and trajectory smoothness and continuity are ensured simultaneously through two-dimensional weighted optimization.

[0046] In some specific embodiments, when using a fifth-order polynomial spline curve to fit the sequence of necessary points to generate a reference path, the optimization direction is to minimize the distance error between path points and necessary points and to minimize the integral of the rate of curvature change along the path. The system solves this objective function and can output a reference path with continuous position, tangential angle, and curvature. The same optimization effect can also be achieved by using an equivalent function form with added higher-order smoothing terms.

[0047] This weighted optimization objective function ensures that the path accurately matches the geometric constraints of the necessary points, and makes the trajectory smoother by minimizing the integral of the rate of change of curvature, thus significantly reducing the steering fluctuations during trajectory tracking.

[0048] As a specific embodiment of this disclosure, based on the basic scheme, it is further defined as follows: the coupling constraint conditions of longitudinal acceleration and lateral acceleration satisfy the following friction ellipse equation:

[0049] in, For longitudinal acceleration, For speed, For curvature, The dynamic adhesion coefficient, It is the acceleration due to gravity. This is the maximum permissible lateral acceleration.

[0050] Specifically, regarding the initial velocity curve planning constraints in step S102 (the self-generation step of limit trajectory and velocity) in the main embodiment, in an optional refined implementation, the coupling restrictions of the vehicle's longitudinal acceleration and lateral acceleration are clearly quantified and limited, and the friction ellipse equation is used as the core physical constraint for velocity planning to ensure that the velocity curve completely fits the vehicle dynamic boundary under the current road surface adhesion.

[0051] In some specific embodiments, when solving for velocity based on the arc length parameter distribution of the reference trajectory, the coupling constraint condition between longitudinal acceleration and lateral acceleration must satisfy the above friction ellipse equation. The system uses this equation as a rigid constraint to solve for velocity, ensuring that the planned initial velocity curve does not exceed the tire grip limit. In addition to the above standard friction ellipse constraint form, an equivalent constraint equation that introduces a load correction coefficient can also be used to achieve the same constraint effect.

[0052] The frictional elliptical coupling constraint equation precisely defines the combined limits of the vehicle's longitudinal and lateral accelerations, preventing speed planning from exceeding the road surface grip range from a physical perspective, thus ensuring that the initial speed curve has absolute feasibility and safety.

[0053] As a specific implementation of this disclosure, based on the basic scheme, a dynamic programming algorithm is used to solve for the optimal speed curve that minimizes the total time to traverse the entire path, which is then used as the initial speed curve. This initial speed curve is further defined as follows: when the confidence level of the instantaneous peak adhesion coefficient is detected to be lower than a preset condition, the optimal speed curve is multiplied by a confidence coefficient less than 1 to reduce the target speed of the first attempt; or, the initial target lateral acceleration is set to a preset proportion of the maximum permissible lateral acceleration to reserve a safety buffer for subsequent iterative exploration.

[0054] Specifically, in one optional refined implementation, a safety margin adaptive correction logic is added to the optimal speed curve obtained by dynamic programming. This reduces the risk of the first test through two independent optional methods, reserving a safety space for subsequent iterative exploration.

[0055] In some specific embodiments, the system uses a dynamic programming algorithm to solve for the optimal speed curve that minimizes the total time to traverse the entire path. This curve is then used as the initial speed curve benchmark and a safety correction is performed. When the confidence level of the instantaneous peak adhesion coefficient is detected to be lower than a preset condition, the optimal speed curve is multiplied by a confidence coefficient less than 1. The speed curve after coefficient scaling is used as the target speed for the first attempt to offset the planning deviation caused by adhesion coefficient detection fluctuations. Alternatively, a lateral acceleration limitation method is used. Instead of directly correcting the speed curve, the initial target lateral acceleration is set to a preset proportion of the maximum allowable lateral acceleration, retaining the physical limit of the corresponding proportion as a safety buffer to adapt to the speed increase requirements of subsequent closed-loop iterations. In addition to the above two correction methods, an equivalent method combining vehicle speed segment constraints can also be used to achieve the same safety buffer effect.

[0056] By adaptively scaling the confidence coefficient or limiting the lateral acceleration ratio, a dynamic safety buffer is embedded for the first test, effectively avoiding vehicle instability caused by overly aggressive speed planning in the initial attempt, and ensuring a smooth start to iterative exploration.

[0057] As a specific implementation of this disclosure, based on the basic scheme, the vehicle is controlled to perform tests according to the reference trajectory and initial speed curve. Based on the comparison results of the real-time monitored vehicle state response and the preset limit criteria, the target speed is automatically adjusted and multiple rounds of iterative exploration are carried out until the limit speed is converged. It is further defined as follows: during the vehicle's driving process, the yaw rate, the center of gravity sideslip angle, and the wheel speed difference are monitored in real time, and the monitored data are compared with the model prediction values; when the deviation between the actual response and the ideal model exceeds the preset threshold or a pile crushing event is detected, the current speed is determined to be the limit point; the highest speed of successful passage and the critical speed of failure are recorded, and the target speed for the next attempt is automatically generated using the bisection method or the golden section method, forming a closed-loop self-learning iterative process until the limit speed is converged.

[0058] Specifically, in one optional refined implementation, the vehicle state monitoring, limit determination rules, and target speed iteration logic are further concretized and defined. By combining multi-dimensional state comparison, quantitative limit criteria, and numerical iteration algorithms, efficient and accurate convergence of the limit passing speed is achieved.

[0059] In some specific embodiments, during the test performed by the vehicle according to the reference trajectory and initial speed curve, the onboard state monitoring module collects real-time vehicle state response data such as yaw rate, center of gravity sideslip angle, and wheel speed difference. This monitoring data is then compared in real-time with the predicted values ​​from the ideal vehicle dynamics model. When the deviation between the actual response data and the ideal model exceeds a preset threshold, or when the sensor detects that the vehicle has run over a cone, the system immediately determines the current test target speed as the limit point. The system simultaneously records the highest speed successfully passed during the test and the critical speed at which the limit is triggered. Based on the bisection method or the golden section method, the speed interval is divided and calculated, automatically generating the target speed for the next test. This closed-loop self-learning iterative process of monitoring, comparison, determination, and adjustment is continuously executed until the target speed stabilizes and converges to the limit speed of the current test condition. Besides the bisection method and the golden section method, other numerical optimization iterative methods can also be used to achieve the same speed convergence effect.

[0060] By comparing vehicle states in real time across multiple dimensions and clarifying limit criteria, the vehicle's limit state can be accurately identified. Combined with the bisection method / golden section method iteration, the convergence efficiency of the limit speed and the stability and repeatability of the test results are significantly improved.

[0061] As a specific implementation of this disclosure, based on the basic scheme, and using the converged limit speed, dynamic adhesion coefficient, and trajectory tracking accuracy during the test process, a normalized objective evaluation score is calculated and generated. This is further specified as follows: the objective evaluation score is calculated using a normalized scoring model, which is:

[0062] in, For the maximum passing speed, The dynamic adhesion coefficient, It is the acceleration due to gravity. The radius of curvature of the working condition is automatically calculated from the geometric boundary. The trajectory fit coefficient is used to reflect the accuracy of a vehicle in tracking an ideal trajectory.

[0063] Specifically, in one optional refined implementation, the calculation model for the objective evaluation score is explicitly quantified and limited, and standardized evaluation results that can be compared horizontally across scenarios are generated by coupling limit speed, road surface adhesion, working condition geometry and trajectory tracking accuracy.

[0064] In some specific embodiments, the system employs a normalized scoring model to calculate the objective evaluation score. The measured and calculated parameters are substituted into the model to complete the numerical calculation, directly outputting the normalized objective evaluation score. Besides the standard scoring model mentioned above, an equivalent model incorporating a working condition correction factor can also achieve the same normalized evaluation effect. This normalized scoring model precisely couples core test parameters, eliminating the influence of road surface adhesion and working condition geometry, ensuring that evaluation scores under different test conditions have a unified quantitative benchmark, and achieving an objective and comparable assessment of vehicle performance limits.

[0065] It should be noted that the embodiments of this disclosure may include multiple steps. For ease of description, these steps are numbered, but these numbers are not a limitation on the execution time slots or execution order between the steps; these steps can be implemented in any order, and the embodiments of this disclosure do not limit this.

[0066] Corresponding to the aforementioned automated vehicle limit performance testing method, this disclosure also proposes an automated vehicle limit performance testing device. Since the device embodiments of this disclosure correspond to the aforementioned method embodiments, details not disclosed in the device embodiments can be referred to the aforementioned method embodiments, and will not be repeated here.

[0067] Figure 2 This is a schematic diagram of the structure of an automated vehicle limit performance testing device provided in an embodiment of this disclosure, as shown below. Figure 2 As shown, it includes: The calculation unit 21 is used to autonomously excite and collect vehicle response data through on-board sensors to calculate the dynamic adhesion coefficient of the current road surface, and at the same time autonomously identify the geometric boundary of the test area to generate a sequence of necessary points for trajectory planning. The generation unit 22 is used to construct a vehicle dynamics constraint model based on the dynamic adhesion coefficient, and to generate a reference trajectory that meets physical feasibility and the initial velocity curve corresponding to the reference trajectory through optimal control by using geometric boundaries and a sequence of necessary points. Test unit 23 is used to control the vehicle to perform tests according to the reference trajectory and initial speed curve. Based on the comparison results of the real-time monitored vehicle state response and the preset limit criteria, it automatically adjusts the target speed and conducts multiple rounds of iterative exploration until it converges to the limit passing speed. The calculation unit 24 is used to calculate and generate a normalized objective evaluation score based on the converged limit pass speed, dynamic adhesion coefficient and trajectory tracking accuracy during the test process.

[0068] It should be noted that the foregoing explanation of the method embodiments also applies to the apparatus of this embodiment, and the principle is the same, so it is not limited in this embodiment.

[0069] According to embodiments of this disclosure, this disclosure also provides an electronic device, a readable storage medium, and a computer program product.

[0070] Figure 3 A schematic block diagram of an example electronic device 300 that can be used to implement embodiments of the present disclosure is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device may also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the present disclosure described and / or claimed herein.

[0071] like Figure 3 As shown, the electronic device 300 includes a computing unit 301, which can perform various appropriate actions and processes based on a computer program stored in ROM (Read-Only Memory) 302 or a computer program loaded from storage unit 308 into RAM (Random Access Memory) 303. The RAM 303 may also store various programs and data required for the operation of the electronic device 300. The computing unit 301, ROM 302, and RAM 303 are interconnected via a bus 304. An I / O (Input / Output) interface 305 is also connected to the bus 304.

[0072] Multiple components in electronic device 300 are connected to I / O interface 305, including: input unit 306, such as keyboard, mouse, etc.; output unit 307, such as various types of displays, speakers, etc.; storage unit 308, such as disk, optical disk, etc.; and communication unit 309, such as network card, modem, wireless transceiver, etc. Communication unit 309 allows electronic device 300 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0073] The computing unit 301 can be various general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the computing unit 301 include, but are not limited to, CPUs (Central Processing Units), GPUs (Graphics Processing Units), various special-purpose AI (Artificial Intelligence) computing chips, various computing units running machine learning model algorithms, DSPs (Digital Signal Processors), and any suitable processor, controller, microcontroller, etc. The computing unit 301 performs the various methods and processes described above, such as automated vehicle performance testing methods. For example, in some embodiments, the automated vehicle performance testing methods can be implemented as computer software programs tangibly contained in a machine-readable medium, such as storage unit 308. In some embodiments, part or all of the computer program can be loaded and / or installed on the electronic device 300 via ROM 302 and / or communication unit 309. When the computer program is loaded into RAM 303 and executed by the computing unit 301, one or more steps of the methods described above can be performed. Alternatively, in other embodiments, the computing unit 301 may be configured to perform the aforementioned automated vehicle limit performance testing method by any other suitable means (e.g., by means of firmware).

[0074] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, FPGAs (Field Programmable Gate Arrays), ASICs (Application-Specific Integrated Circuits), ASSPs (Application-Specific Standard Products), SOCs (System-on-Chips), CPLDs (Complex Programmable Logic Devices), computer hardware, firmware, software, and / or combinations thereof. These various implementations may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0075] The program code used to implement the methods of this disclosure may be written in any combination of one or more programming languages. This program code may be provided to a processor or controller of a general-purpose computer, special-purpose computer, or other programmable data processing apparatus, such that when executed by the processor or controller, the program code causes the functions / operations specified in the flowcharts and / or block diagrams to be implemented. The program code may be executed entirely on a machine, partially on a machine, as a standalone software package partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0076] In the context of this disclosure, a machine-readable medium can be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can be, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, RAM, ROM, EPROM (Electrically Programmable Read-Only Memory) or flash memory, optical fiber, CD-ROM (Compact Disc Read-Only Memory), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0077] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device for displaying information to the user (e.g., a CRT (Cathode-Ray Tube) or LCD (Liquid Crystal Display) monitor); and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the computer. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0078] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or computing systems that include middleware components (e.g., application servers), or computing systems that include frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include LANs (Local Area Networks), WANs (Wide Area Networks), the Internet, and blockchain networks.

[0079] Computer systems can include clients and servers. Clients and servers are generally geographically separated and typically interact via communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. A server can be a cloud server, also known as a cloud computing server or cloud host, a hosting product within the cloud computing service system that addresses the shortcomings of traditional physical hosts and VPS (Virtual Private Server) services, such as high management difficulty and weak business scalability. Servers can also be servers for distributed systems or servers incorporating blockchain technology.

[0080] It's important to note that artificial intelligence (AI) is the study of enabling computers to simulate certain human thought processes and intelligent behaviors (such as learning, reasoning, thinking, and planning). It encompasses both hardware and software technologies. AI hardware technologies generally include sensors, dedicated AI chips, cloud computing, distributed storage, and big data processing. AI software technologies primarily include computer vision, speech recognition, natural language processing, machine learning / deep learning, big data processing, and knowledge graph technologies.

[0081] The various numerical designations such as "first," "second," etc., used in this disclosure are merely for ease of description and are not intended to limit the scope of the embodiments of this disclosure, nor do they indicate a sequential order.

[0082] At least one of the features described in this disclosure can also be described as one or more, and multiple features can be two, three, four or more, and this disclosure does not impose any limitations. In the embodiments of this disclosure, for a technical feature, the technical features in that technical feature are distinguished by "first", "second", "third", "A", "B", "C" and "D", etc., and there is no sequential order or size order among the technical features described by "first", "second", "third", "A", "B", "C" and "D".

[0083] It should be understood that the various forms of processes shown above can be used to rearrange, add, or delete steps. For example, the steps described in this disclosure can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution disclosed in this disclosure can be achieved, and this is not limited herein.

[0084] The specific embodiments described above do not constitute a limitation on the scope of protection of this disclosure. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this disclosure should be included within the scope of protection of this disclosure.

Claims

1. An automated method for testing the extreme performance of vehicles, characterized in that, include: The vehicle autonomously excites and collects vehicle response data through onboard sensors to calculate the dynamic adhesion coefficient of the current road surface. At the same time, it autonomously identifies the geometric boundaries of the test area and generates a sequence of necessary points for trajectory planning. A vehicle dynamics constraint model is constructed based on the dynamic adhesion coefficient, and a reference trajectory that satisfies physical feasibility and an initial velocity curve corresponding to the reference trajectory are generated through optimal control by utilizing the geometric boundary and the necessary point sequence. The vehicle is controlled to perform the test according to the reference trajectory and the initial speed curve. Based on the comparison results of the real-time monitored vehicle state response and the preset limit criterion, the target speed is automatically adjusted and multiple rounds of iterative exploration are carried out until the limit speed is converged. Based on the converged limit speed, the dynamic adhesion coefficient, and the trajectory tracking accuracy during the test, a normalized objective evaluation score is calculated and generated.

2. The method according to claim 1, characterized in that, The process of autonomously exciting and collecting vehicle response data through onboard sensors to calculate the dynamic adhesion coefficient of the current road surface, and simultaneously autonomously identifying the geometric boundaries of the test area to generate a sequence of necessary points for trajectory planning, includes: The steering system autonomously applies a sinusoidal sweep frequency input or the braking system applies a micro-braking pulse as an excitation signal, and the vehicle's lateral acceleration or longitudinal deceleration is collected as vehicle response data; The instantaneous peak adhesion coefficient is calculated in real time by comparing the excitation signal with the vehicle response data using the recursive least squares method. By using vehicle-mounted vision or point cloud sensors to identify the characteristics of the traffic cones and establish a traffic cone point cloud, the system can automatically distinguish different functional areas or identify the axis of the traffic cone arrangement, calculate the spacing between traffic cones, lane width, and entrance / exit location information in real time, and generate a sequence of necessary points that vehicles must pass through.

3. The method according to claim 2, characterized in that, The process of constructing a vehicle dynamics constraint model based on the dynamic adhesion coefficient, and using the geometric boundaries and the sequence of necessary points to generate a physically feasible reference trajectory and an initial velocity curve corresponding to the reference trajectory through optimal control solution, includes: A friction ellipse model is constructed based on the instantaneous peak adhesion coefficient, and the maximum allowable lateral acceleration and velocity-curvature constraint relationship of any path point under a given path curvature are derived. The necessary point sequence is adaptively shrunk according to the instantaneous peak adhesion coefficient. When the instantaneous peak adhesion coefficient is lower than a preset threshold, the necessary points are shrunk inward toward the center line of the channel. A fifth-order polynomial spline curve is used to fit the sequence of necessary points after adaptive shrinkage. The optimization objective is to minimize the distance error between the path points and the necessary points and the integral of the rate of curvature change along the path, thereby generating a reference path with continuous position, tangential angle and curvature. Based on the coupling constraints of longitudinal acceleration and lateral acceleration defined by the friction ellipse model, a dynamic programming algorithm is used to solve for the optimal velocity curve that minimizes the total time to traverse the entire path, which is then used as the initial velocity curve.

4. The method according to claim 3, characterized in that, The optimization objective function, which aims to minimize the distance error between path points and necessary points, as well as the integral of the rate of curvature change along the path, is as follows: in, For path points, As a necessary point, For curvature, Let the arc length be , and These are the weighting coefficients.

5. The method according to claim 3, characterized in that, The coupling constraint between the longitudinal acceleration and the lateral acceleration satisfies the following friction ellipse equation: in, For longitudinal acceleration, For speed, For curvature, The dynamic adhesion coefficient is... It is the acceleration due to gravity. This refers to the maximum permissible lateral acceleration.

6. The method according to claim 3, characterized in that, The step of using dynamic programming to find the optimal speed curve that minimizes the total time to traverse the entire path, and using it as the initial speed curve, includes: When the confidence level of the instantaneous peak adhesion coefficient is detected to be lower than a preset condition, the optimal speed curve is multiplied by a confidence level less than 1 to reduce the target speed of the first attempt; or, The initial target lateral acceleration is set to a preset proportion of the maximum permissible lateral acceleration to reserve a safety buffer for subsequent iterative exploration.

7. The method according to claim 1, characterized in that, The controlled vehicle performs the test according to the reference trajectory and the initial speed curve. Based on the comparison results of the real-time monitored vehicle state response and the preset limit criterion, it automatically adjusts the target speed and conducts multiple rounds of iterative exploration until it converges to the limit passing speed, including: The yaw rate, center of gravity sideslip angle, and wheel speed difference are monitored in real time during vehicle operation, and the monitoring data are compared with the model prediction values. When the deviation between the actual response and the ideal model exceeds a preset threshold or a pile compaction event is detected, the current speed is determined to be the limit point. Record the highest speed achieved on successful attempts and the critical speed at which failures occur. Use the bisection method or the golden section method to automatically generate the target speed for the next attempt, forming a closed-loop self-learning iterative process until convergence to the aforementioned limit speed.

8. The method according to claim 1, characterized in that, The normalized objective evaluation score is calculated and generated based on the converged limit speed, the dynamic adhesion coefficient, and the trajectory tracking accuracy during the test, including: The objective evaluation score is calculated using a normalized scoring model, which is as follows: in, The limiting speed is... The dynamic adhesion coefficient is... It is the acceleration due to gravity. The radius of curvature of the working condition is automatically calculated from the geometric boundary. The trajectory fit coefficient is used to reflect the accuracy of a vehicle in tracking an ideal trajectory.

9. An automated testing device for vehicle extreme performance, characterized in that, include: The calculation unit is used to autonomously excite and collect vehicle response data through onboard sensors to calculate the dynamic adhesion coefficient of the current road surface, and at the same time autonomously identify the geometric boundary of the test area to generate a sequence of necessary points for trajectory planning. The generation unit is used to construct a vehicle dynamics constraint model based on the dynamic adhesion coefficient, and use the geometric boundary and the sequence of necessary points to generate a reference trajectory that satisfies physical feasibility and an initial velocity curve corresponding to the reference trajectory through optimal control solution. The testing unit is used to control the vehicle to perform tests according to the reference trajectory and the initial speed curve. Based on the comparison results of the real-time monitored vehicle state response and the preset limit criterion, it automatically adjusts the target speed and performs multiple rounds of iterative exploration until it converges to the limit passing speed. The calculation unit is used to calculate and generate a normalized objective evaluation score based on the converged limit passing speed, the dynamic adhesion coefficient, and the trajectory tracking accuracy during the test.

10. An electronic device, characterized in that, include: At least one processor; as well as A memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor to enable the at least one processor to perform the method of any one of claims 1-8.

Images