A road driving adaptability evaluation method for autonomous driving vehicles based on virtual simulation

Abstract

The present application relates to a method for evaluating the road driving adaptability of an autonomous driving vehicle based on virtual simulation. The method includes: obtaining the driving condition information of the autonomous driving vehicle; using the vehicle driving condition information to construct an autonomous driving vehicle simulation test scene in the simulation software and conduct a test to obtain the obtainable sight distance; establishing an autonomous driving sight distance failure function function based on the required parking sight distance and the obtainable sight distance of different autonomous driving levels; analyzing the autonomous driving sight distance failure probability of each position point on the target road based on the autonomous driving sight distance failure function to obtain the autonomous driving sight distance failure probability of each position point; analyzing the segmentation results of the target road and the autonomous driving sight distance failure probability of each position point to determine the adaptability evaluation results of each segment of the autonomous driving vehicle along the target road. It can efficiently and safely quantify and evaluate the driving adaptability of the autonomous driving vehicle on the target road.

Description

Technical Field

[0001] The present application relates to the field of road driving technology, and in particular to a method for evaluating the road driving adaptability of an autonomous driving vehicle based on virtual simulation. Background Art

[0002] Autonomous driving vehicles, with intelligence as an important feature, are bringing major changes to the automotive industry. Autonomous driving technology is expected to help improve road safety, traffic efficiency, ease traffic congestion, reduce driving energy consumption and reduce environmental pollution. The autonomous driving system, which uses environmental perception technology, intelligent decision-making technology and control execution technology as the main technical architecture, uses machine vision and radar to replace the driver's vision; uses on-board computing units, decision-making units and control units to replace the driver's brain; and uses motion actuators such as brakes, accelerators, steering, and speed changes to replace the driver's limbs. With the improvement of the level of autonomous driving, the "people" in the original closed-loop traffic system of man-vehicle-road will gradually transform into a "system". Therefore, the difference between the autonomous driving system and the driver's perception mode is the source of the change in the vehicle-road coupling relationship in the two traffic systems.

[0003] At present, accidents involving mass-produced low-level autonomous driving vehicles occur frequently, while high-level autonomous driving still cannot completely drive safely and reliably on open roads without safety personnel. Drivers' blind trust in the performance of autonomous driving perception technology in real road environments has become one of the main reasons for the above accidents. However, there are few studies in the industry that discuss the issue of "whether autonomous driving vehicles are suitable for existing road infrastructure and roadside facilities designed and built for manned vehicles." Therefore, how to efficiently, safely and effectively evaluate the driving adaptability of autonomous driving vehicles on existing roads has become a key issue that needs to be solved urgently to ensure the safe driving of autonomous driving. Summary of the invention

[0004] Based on this, it is necessary to provide a road driving adaptability evaluation method for autonomous driving vehicles based on virtual simulation, which can efficiently, safely and effectively evaluate the driving adaptability of autonomous driving vehicles on existing roads in response to the above technical problems.

[0005] A method for evaluating the road driving adaptability of an autonomous driving vehicle based on virtual simulation, the method comprising:

[0006] Acquire driving condition information of the autonomous driving vehicle, wherein the driving condition information includes road design information, autonomous driving vehicle information, and roadside and mid-road environment information;

[0007] Using the vehicle driving condition information, a simulation test scenario for an autonomous driving vehicle is constructed in a simulation software and tested to obtain an obtainable sight distance;

[0008] Based on the required parking sight distances of different autonomous driving levels and the obtainable sight distances, an autonomous driving sight distance failure function is established;

[0009] Analyzing the probability of automatic driving sight distance failure at each position point of the target road based on the automatic driving sight distance failure function to obtain the probability of automatic driving sight distance failure at each position point;

[0010] An analysis is performed based on the segmentation results of the target road and the probability of failure of the autonomous driving sight range at each of the location points to determine the adaptability evaluation results of each segment of the autonomous driving vehicle traveling along the target road.

[0011] In one embodiment, when the autonomous driving level is L1 or L2 autonomous driving, the required parking sight distance is:

[0012]

[0013] Among them, S D,ADAS (s) is the required stopping sight distance for L1 or L2 autonomous driving that varies with the position on the driving path (m); V is the driving speed (km / h); t ADAS is the driver assistance system perception reaction time (s); t H A is the response time of L1 and L2 drivers taking over (s); d is the driver's braking deceleration (m / s 2 );i L (s) is the longitudinal slope of the road that changes with the position point on the driving path; it takes + when the road line is uphill and - when it is downhill.

[0014] In one embodiment, when the autonomous driving level is L3 autonomous driving, the required parking sight distance is:

[0015]

[0016] Among them, S D,L3 (s) is the required parking sight distance for L3 autonomous driving that changes with the position point on the driving path (m); t ADS is the perception-braking reaction time of the autonomous driving system (s); t T The response time for the L3 driver to take over (s); A dp Preset braking deceleration for L3 autonomous driving system (m / s 2 ); when the road line is uphill, take +; when it is downhill, take -.

[0017] In one embodiment, when the autonomous driving level is L4 or L5 autonomous driving, the required parking sight distance is:

[0018]

[0019] Among them, S D,ADS (s) is the required stopping sight distance (m) for L4 or L5 autonomous driving that varies with the position point on the driving path; A ds is the braking deceleration of the autonomous driving system (m / s 2 ); when the road line is uphill, take +; when it is downhill, take -.

[0020] In one embodiment, the autonomous driving sight range failure function Z S (s) is:

[0021]

[0022] Among them, S A,ADAS (s) is the available visual distance for L1 or L2 autonomous driving (m); S A,L3 (s) is the available visual distance for L3 autonomous driving (m); S A,ADS (s) is the available visual distance (m) for L4 or L5 autonomous driving.

[0023] In one embodiment, the step of analyzing the probability of automatic driving sight distance failure at each position point of the target road based on the automatic driving sight distance failure function to obtain the probability of automatic driving sight distance failure at each position point includes:

[0024] Obtain variable characteristic values ​​of required stopping sight distances for different levels of autonomous driving at various locations on the target road;

[0025] According to the variable characteristic value and the obtainable sight distance, the automatic driving sight distance failure function is solved by using a first-order second moment method for solving reliability to obtain the automatic driving sight distance failure probability at each position point.

[0026] In one embodiment, the step of analyzing the segmentation result of the target road and the probability of failure of the autonomous driving sight range at each position point to determine the adaptability evaluation result of each segment of the autonomous driving vehicle traveling along the target road includes:

[0027] Based on the adaptability index calculation formula, the adaptability index of each of the position points is determined by analyzing the probability of failure of the autonomous driving sight range at each of the position points;

[0028] Segmenting the target road according to a preset segmentation method to obtain a segmentation result of the target road;

[0029] Analyze the adaptability index of the position points corresponding to each segment of the target road segmentation result to obtain the average adaptability index corresponding to each segment of the target road;

[0030] According to the adaptability level range to which the average value of the adaptability index corresponding to each segment of the target road belongs, the adaptability evaluation results of each segment along which the automatic driving vehicle travels along the target road are determined.

[0031] In one embodiment, the adaptability level range includes:

[0032] When 0≤average value of adaptability index<0.25, the adaptability level is poor;

[0033] When 0.25≤average value of adaptability index<0.50, the adaptability level is poor;

[0034] When 0.50≤average value of adaptability index<0.75, the adaptability level is good;

[0035] When 0.75≤average value of adaptability index≤1, the adaptability level is good.

[0036] In one embodiment, the adaptability index calculation formula is:

[0037] β(s)=1-p f (s)

[0038] Among them, β(s) is the adaptability index of position point s, p f (s) is the probability of failure of autonomous driving sight range at position s.

[0039] The above-mentioned road driving adaptability evaluation method of the autonomous driving vehicle based on virtual simulation obtains the driving condition information of the autonomous driving vehicle, and the driving condition information includes road design information, autonomous driving vehicle information and roadside and road environment information; uses the vehicle driving condition information to construct an autonomous driving vehicle simulation test scene in the simulation software and conducts a test to obtain the obtainable sight distance; establishes an autonomous driving sight distance failure function function based on the required parking sight distance of different autonomous driving levels and the obtainable sight distance; analyzes the autonomous driving sight distance failure probability of each position point of the target road based on the autonomous driving sight distance failure function to obtain the autonomous driving sight distance failure probability of each position point; and analyzes the segmentation result of the target road and the autonomous driving sight distance failure probability of each position point to determine the adaptability evaluation results of each segment of the autonomous driving vehicle traveling along the target road. It can efficiently and safely quantify and evaluate the driving adaptability of the autonomous driving vehicle on the target road, identify the dangerous road sections where the autonomous driving vehicle has insufficient sight distance, provide theoretical reference for the design and optimization of the warning function of the autonomous driving system, and also conduct safety hazard inspection for the function test of the autonomous driving system and the actual operation of the existing road infrastructure. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 1 is a flow chart of a method for evaluating the road driving adaptability of an autonomous driving vehicle based on virtual simulation in one embodiment;

[0041] Figure 2 is a schematic diagram of the corresponding relationship between the elements of the autonomous driving vehicle simulation test scenario and the driving condition information of the autonomous driving vehicle in an embodiment;

[0042] Figure 3 The figure is a top view of a simulation test scene of an autonomous driving vehicle in an embodiment, wherein 1 is the autonomous driving vehicle, 2 is the left road boundary, 3 is the horizontal field of view of the perception sensor, 4 is the right road boundary, and 5 is a roadside obstacle. DETAILED DESCRIPTION

[0043] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0044] In one embodiment, Figure 1 As shown, a method for evaluating the road driving adaptability of an autonomous driving vehicle based on virtual simulation is provided, comprising the following steps:

[0045] Step S220, obtaining driving condition information of the autonomous driving vehicle, the driving condition information including road design information, autonomous driving vehicle information, and roadside and mid-road environment information.

[0046] The road design information includes a table of straight lines, curves and corners along the road pile number, a vertical curve design table, a roadbed design table, and a cross-section design drawing. Among them, the road straight line, curve and corner table information includes intersection number, intersection stake number and intersection coordinates, intersection spacing, calculated azimuth, straight line length between curves, corner, curve element table and curve stationary point position. The curve element table includes tangent length, radius, relaxation parameter, curve length, total curve length and external distance; the vertical curve design table includes slope change point sequence number, slope change point stake number, elevation, longitudinal slope, slope length, vertical curve element and curve position and straight slope section length. The vertical curve element and curve position include break, concave vertical curve radius, convex vertical curve radius, vertical curve length, vertical curve external distance, vertical curve starting stake number, vertical curve ending stake number; the roadbed design table includes stake number, ground elevation, design elevation, fill and cut height, distance and elevation difference between each point on the left and right of the cross section and the design line; the cross-section design drawing includes stake number, number of lanes, lane width, shoulder width, central dividing strip width, curb width, slope length and slope, road cross slope, and superelevation.

[0047] The autonomous driving vehicle information includes the autonomous driving level, vehicle model, perception sensor layout plan and design parameters, among which the autonomous driving level includes L1 to L5; the vehicle type includes passenger cars and trucks; the perception sensor layout plan includes the type, quantity, installation height and position of perception sensors; the design parameters include the plane field of view angle, vertical field of view angle, maximum field of view distance, plane field of view resolution, and vertical field of view resolution.

[0048] The roadside and mid-road environmental information includes green belts, traffic signs and building layout information along the road pile number changes, among which the building layout information includes location and height.

[0049] Step S240, using the vehicle driving condition information to construct an autonomous driving vehicle simulation test scenario in the simulation software and perform a test to obtain an obtainable sight distance.

[0050] Among them, the simulation software mainly focuses on road scene construction, perception sensor model and vehicle dynamics model simulation, which is used to simulate the available sight distance of autonomous driving vehicles in road scenes, such as CarSim, TruckSim (vehicle dynamics model simulation software), SCANeR, Prescan, etc. The simulation software of this application is preferably Prescan, because it has a more realistic physics-based sensor model and can provide feedback on changes in obstacle recognition rate in different weather environments. The elements of the autonomous driving vehicle simulation test scene and their corresponding relationship with the driving condition information of the autonomous driving vehicle are as follows: Figure 2 shown.

[0051] Determine the elements of the autonomous driving vehicle simulation test scene, including road scenes, autonomous driving vehicle models, autonomous driving vehicle driving information and sight distance calculation obstacle vehicles, where the road scene includes road line shape and road center and road side environment, and autonomous driving vehicle driving information includes driving speed and driving path. Create a virtual simulation road scene based on the road design information and road center and road side environment information in the vehicle driving condition information, so that the road design information and road center and road side environment in the simulation environment are consistent with the real environment, and set the weather environment in the simulation software; create a virtual autonomous driving vehicle model based on the autonomous driving vehicle information in the vehicle driving condition information and set the vehicle dynamics model in the simulation software, so that the autonomous driving vehicle in the simulation environment is consistent with the real environment; set the autonomous driving vehicle driving information in the simulation software; set the minimum interval, and arrange sight distance calculation obstacle vehicles in the same direction along the vehicle driving path in the simulation software at the minimum interval, where the minimum interval should meet the requirements of the solution to obtain the sight distance, usually set to 2m or 5m; use the simulation software to construct an overhead view of the autonomous driving vehicle simulation test scene as shown below Figure 3 shown.

[0052] A test experiment is conducted. According to the farthest sight distance recognized by the perception sensor that changes with the position point on the driving path obtained in the simulation software, the position point of the obstacle vehicle on the driving path is calculated, and the driving path difference between the position point of the obstacle vehicle on the driving path and the position point of the autonomous driving vehicle on the driving path is calculated as the obtainable sight distance S that changes with the position point on the driving path. A (s), where s is the position of the autonomous vehicle on the driving path. Take the determination of the obtainable sight distance at time T during testing in the simulation software as an example: set the farthest sight distance output by the perception sensor at each simulation time in the simulation software, determine the position of the obstacle vehicle on the driving path at each simulation time, set the position of the autonomous driving vehicle on the driving path at each simulation time, match the farthest sight distance at simulation time T, calculate the position of the obstacle vehicle on the driving path and the position of the autonomous driving vehicle on the driving path, and calculate the position of the obstacle vehicle on the driving path - the position of the autonomous driving vehicle on the driving path = the obtainable sight distance S. A (s), and obtain the available viewing distance at time T.

[0053] Step S260: Establishing an automatic driving sight distance failure function based on the required parking sight distance and the obtainable sight distance of different automatic driving levels.

[0054] The required parking sight distance for different levels of autonomous driving is determined according to the following formula. When the autonomous driving level is L1 or L2, the required parking sight distance is:

[0055]

[0056] Among them, S D,ADAS (s) is the required stopping sight distance for L1 or L2 autonomous driving that varies with the position on the driving path (m); V is the driving speed (km / h); t ADAS is the driver assistance system perception reaction time (s); t H A is the response time of L1 and L2 drivers taking over (s); d is the driver's braking deceleration (m / s 2 );i L (s) is the longitudinal slope of the road that changes with the position point on the driving path; it takes + when the road line is uphill and - when it is downhill.

[0057] When the autonomous driving level is L3, the required parking sight distance is:

[0058]

[0059] Among them, S D,L3(s) is the required parking sight distance for L3 autonomous driving that changes with the position point on the driving path (m); t ADS is the perception-braking reaction time of the autonomous driving system (s); t T The response time for the L3 driver to take over (s); A dp Preset braking deceleration for L3 autonomous driving system (m / s 2 ); when the road line is uphill, take +; when it is downhill, take -.

[0060] When the autonomous driving level is L4 or L5, the required parking sight distance is:

[0061]

[0062] Among them, S D,ADS (s) is the required stopping sight distance (m) for L4 or L5 autonomous driving, which varies with the position point on the driving path; A ds is the braking deceleration of the autonomous driving system (m / s 2 ); when the road line is uphill, take +; when it is downhill, take -.

[0063] Based on the required parking sight distance and available sight distance of different autonomous driving levels, the autonomous driving sight distance failure function Z is established S (s) is:

[0064]

[0065] Among them, S A,ADAS (s) is the available visual range for L1 or L2 autonomous driving (m); S A,L3 (s) is the available visual distance for L3 autonomous driving (m); S A,ADS (s) is the available visual distance (m) for L4 or L5 autonomous driving.

[0066] Step S280, analyzing the probability of automatic driving sight distance failure at each position point of the target road based on the automatic driving sight distance failure function, and obtaining the probability of automatic driving sight distance failure at each position point.

[0067] In one embodiment, the step of analyzing the probability of autonomous driving sight distance failure at each position point of the target road based on the autonomous driving sight distance failure function to obtain the probability of autonomous driving sight distance failure at each position point includes: obtaining variable characteristic values ​​of required parking sight distances of different autonomous driving levels at each position point of the target road; solving the autonomous driving sight distance failure function function using a first-order second moment method for solving reliability according to the variable characteristic values ​​and the obtainable sight distance, to obtain the probability of autonomous driving sight distance failure at each position point.

[0068] Among them, the variable characteristic values ​​include driving speed, road longitudinal slope gradient, L3 level automatic driving system preset braking deceleration, driver braking deceleration, automatic driving system braking deceleration, as well as the distribution form, mean and standard deviation of driving assistance system perception reaction time, L1 level driver takeover reaction time, L2 level driver takeover reaction time, L3 level driver takeover reaction time and automatic driving system perception-braking reaction time; because driving speed, road longitudinal slope gradient, L3 level automatic driving system preset braking deceleration, driver braking deceleration, automatic driving system braking deceleration are deterministic variables, driving speed, road longitudinal slope gradient, L3 level automatic driving system preset braking deceleration, driver braking deceleration, automatic driving system braking deceleration are directly used as variable characteristic values; because driving assistance system perception reaction time, L1 level driver takeover reaction time, L2 level driver takeover reaction time, L3 level driver takeover reaction time and automatic driving system perception-braking reaction time are random variables, driving assistance system perception reaction time, L1 level driver takeover reaction time, The distribution form, mean and standard deviation of L2 driver takeover reaction time, L3 driver takeover reaction time and autonomous driving system perception-braking reaction time are used as variable characteristic values.

[0069] The first-order second-moment method is to expand the nonlinear autonomous driving sight range failure function into a Taylor series and take the first-order term, such as the center point method and the design verification point method. The simulation software of this application preferably uses the design verification point method because it can consider the actual distribution form of random variables.

[0070] Step S300, analyzing the segmentation results of the target road and the probability of failure of the autonomous driving sight range at each location point, and determining the adaptability evaluation results of each segment of the autonomous driving vehicle traveling along the target road.

[0071] In one embodiment, the step of analyzing the segmentation result of the target road and the probability of failure of the autonomous driving sight range at each position point to determine the adaptability evaluation result of each segment of the autonomous driving vehicle traveling along the target road includes: determining the adaptability index of each position point based on the adaptability index calculation formula and analyzing the probability of failure of the autonomous driving sight range at each position point; segmenting the target road according to a preset segmentation method to obtain the segmentation result of the target road; analyzing the adaptability index of the position points corresponding to each segment according to the segmentation result of the target road to obtain the average value of the adaptability index corresponding to each segment of the target road; and determining the adaptability evaluation result of each segment of the autonomous driving vehicle traveling along the target road according to the adaptability level range to which the average value of the adaptability index corresponding to each segment of the target road belongs.

[0072] Among them, the preset segmentation method can be equidistant segmentation or segmentation by linear attributes. The target road is segmented by equidistant segmentation, such as 20m, 50m, and segmented by linear attributes, which are divided into straight lines, gentle curves, circular curves, and horizontal and vertical combined lines. When the target road has good accident conditions, equidistant segmentation can be used, which is convenient and fast; when the target road has a large number of accidents, segmentation by linear attributes can be used, which helps to take countermeasures against dangerous sections to prevent accident risks.

[0073] In one embodiment, the adaptability level range includes: when 0≤adaptability index average value<0.25, the adaptability level is poor; when 0.25≤adaptability index average value<0.50, the adaptability level is relatively poor; when 0.50≤adaptability index average value<0.75, the adaptability level is relatively good; when 0.75≤adaptability index average value≤1, the adaptability level is good. The adaptability index calculation formula is:

[0074] β(s)=1-p f (s)

[0075] Among them, β(s) is the adaptability index of position point s, p f (s) is the probability of failure of autonomous driving sight range at position s.

[0076] In summary, this application aims at the visual difference between the autonomous driving perception system and the human driver of a manned vehicle, builds an autonomous driving vehicle simulation test scene based on a virtual simulation platform and conducts tests, obtains the available sight distance of the autonomous driving vehicle on the target road, and on this basis, combines the required parking sight distance, uses reliability theory to establish the autonomous driving sight distance failure function function, and calculates the autonomous driving sight distance failure probability based on the test results, and finally calculates the autonomous driving vehicle's driving adaptability evaluation results on the target road. The method disclosed in this application can be used to efficiently and safely quantify and evaluate the driving adaptability of the autonomous driving vehicle on the target road, identify dangerous sections where the autonomous driving vehicle has insufficient sight distance, provide theoretical references for the design and optimization of the warning function of the autonomous driving system, and also conduct safety hazard inspections for the functional testing of the autonomous driving system and the actual operation of existing road infrastructure.

[0077] It should be understood that although Figure 1 The steps in the flowchart are shown in sequence as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified in this document, there is no strict order restriction for the execution of these steps, and these steps can be executed in other orders. Moreover, Figure 1At least part of the steps may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily executed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed in turn or alternately with other steps or at least part of the sub-steps or stages of other steps.

[0078] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0079] The above-mentioned embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application shall be subject to the attached claims.

Claims

1. A method for evaluating the road driving adaptability of an autonomous driving vehicle based on virtual simulation, characterized in that: The method comprises: Acquire driving condition information of the autonomous driving vehicle, wherein the driving condition information includes road design information, autonomous driving vehicle information, and roadside and mid-road environment information; Using the vehicle driving condition information, a simulation test scenario for an autonomous driving vehicle is constructed in a simulation software and tested to obtain an obtainable sight distance; Based on the required parking sight distances of different autonomous driving levels and the obtainable sight distances, an autonomous driving sight distance failure function is established; Analyzing the probability of automatic driving sight distance failure at each position point of the target road based on the automatic driving sight distance failure function to obtain the probability of automatic driving sight distance failure at each position point; Analyzing the segmentation results of the target road and the probability of failure of the autonomous driving sight range at each of the position points, and determining the adaptability evaluation results of each segment of the autonomous driving vehicle traveling along the target road; When the autonomous driving level is L1 or L2, the required parking sight distance is: Where s is the location point on the driving path; S D,ADAS (s) is the required stopping sight distance for L1 or L2 autonomous driving that varies with the position point on the driving path; V is the driving speed; t ADAS is the perception reaction time of the driving assistance system; t H The response time for L1 and L2 drivers to take over; A d The braking deceleration for the driver; i L (s) is the longitudinal slope of the road that changes with the position point on the driving path; it is + when the road line is uphill and - when it is downhill; When the autonomous driving level is L3 autonomous driving, the required parking sight distance is: Among them, S D,L3 (s) is the required parking sight distance for L3 autonomous driving that changes with the position point on the driving path; t ADS is the perception-braking reaction time of the autonomous driving system; t T Take over the reaction time for the L3 driver; A dp Preset braking deceleration for L3 autonomous driving system; set to + when the road line is uphill and - when it is downhill; When the autonomous driving level is L4 or L5, the required parking sight distance is: Among them, S D,ADS (s) is the required stopping sight distance for L4 or L5 autonomous driving that varies with the position point on the driving path; A ds It is the braking deceleration of the automatic driving system; when the road line is uphill, it is +, and when it is downhill, it is -; The automatic driving sight range failure function Z S (s) is: Among them, S A,ADAS (s) is the available sight distance for L1 or L2 autonomous driving that varies with the position point on the driving path; S A,L3 (s) is the available sight distance for L3 autonomous driving that changes with the position point on the driving path; S A,ADS (s) is the available sight distance for L4 or L5 autonomous driving that varies with the position point on the driving path; The step of analyzing the automatic driving sight distance failure probability of each position point of the target road based on the automatic driving sight distance failure function to obtain the automatic driving sight distance failure probability of each position point includes: Obtain variable characteristic values ​​of required stopping sight distances for different levels of autonomous driving at various locations on the target road; According to the variable characteristic value and the obtainable sight distance, the automatic driving sight distance failure function is solved by using a first-order second moment method for solving reliability to obtain the automatic driving sight distance failure probability at each position point.

2. The method according to claim 1, characterized in that The step of analyzing the segmentation result of the target road and the probability of failure of the autonomous driving sight range at each position point to determine the adaptability evaluation result of each segment of the autonomous driving vehicle traveling along the target road includes: Based on the adaptability index calculation formula, the adaptability index of each of the position points is determined by analyzing the probability of failure of the autonomous driving sight range at each of the position points; Segmenting the target road according to a preset segmentation method to obtain a segmentation result of the target road; Analyzing the adaptability indexes of the position points corresponding to each segment according to the segmentation result of the target road to obtain the average adaptability indexes corresponding to each segment of the target road; According to the adaptability level range to which the average value of the adaptability index corresponding to each segment of the target road belongs, the adaptability evaluation results of each segment along which the automatic driving vehicle travels along the target road are determined.

3. The method according to claim 2, characterized in that The adaptability level range includes: When 0≤average value of adaptability index<0.25, the adaptability level is poor; When 0.25≤average value of adaptability index<0.50, the adaptability level is poor; When 0.50≤average value of adaptability index<0.75, the adaptability level is good; When 0.75≤average value of adaptability index≤1, the adaptability level is good.

4. The method according to claim 2, characterized in that The calculation formula of the adaptability index is: β(s)=1-p f (s) Among them, β(s) is the adaptability index of position point s, p f (s) is the probability of automatic driving sight range failure at position s.

Images