A processing method for evaluating a lane-changing path by a simulation platform

Abstract

This invention relates to a method for evaluating lane-change paths using a simulation platform. The method includes: initializing the simulation platform to evaluate a road scenario; setting the lane-change start point, longitudinal vehicle speed, driver reaction time, and safe lane-change distance; and confirming the lane-change end point; calling a lane-change path simulator to simulate the driving path from the lane-change start point to the lane-change end point, generating a first simulated path; loading multiple path planning models to be evaluated one by one; calling the currently loaded path planning model to plan the driving path from the lane-change start point to the lane-change end point, generating a first planned path; evaluating the path similarity between the first simulated path and each first planned path; forming first model evaluation data composed of the model name of each path planning model and its corresponding first and second similarities; and forming a first evaluation report composed of all the obtained first model evaluation data. This invention can improve path planning performance.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data processing, in particular to a processing method for evaluating a lane-changing path by a simulation platform. BACKGROUND

[0002] When an automatic driving system plans a lane-changing path, a path planning model is used for planning processing, and the performance of the path planning model directly affects the safety of vehicle driving. There are more than one implementation methods of the path planning model in the conventional case, but there is no general implementation scheme for evaluating the performance of multiple alternative path planning models. SUMMARY

[0003] The present application aims to solve the defects of the prior art, and provides a processing method for evaluating a lane-changing path by a simulation platform, an electronic device and a computer readable storage medium. The simulation platform is used to evaluate and provide a lane-changing path simulator for the simulation platform. The simulation platform creates and sets an evaluation road scene before evaluation. The simulation platform calls the lane-changing path simulator to plan a path under the current evaluation road scene to obtain a corresponding simulation path, and then calls each path planning model to be evaluated to plan a path to obtain a corresponding planning path. The first and second similarities are obtained by evaluating the curve spacing and curve shape of each planning path with reference to the simulation path. Finally, an evaluation report is generated according to the first and second similarities. The general scheme for evaluating the performance of multiple alternative path planning models can improve the selection ability of the automatic driving system for the path planning model, improve the path planning performance of the automatic driving system, and improve the safety driving guarantee ability of the automatic driving system.

[0004] To achieve the above-mentioned purpose, the first aspect of the present application provides a processing method for evaluating a lane-changing path by a simulation platform, which comprises:

[0005] The simulation platform initializes an evaluation road scene; the evaluation road scene comprises a starting lane and an ending lane;

[0006] The center line starting point of the lane center line of the starting lane is set as a lane-changing starting point p s , and the longitudinal vehicle speed v0, the driver reaction time t0 and the safe lane-changing distance d0 are set, and the corresponding lane-changing ending point p s is determined according to the lane-changing starting point p e , the longitudinal vehicle speed v0, the driver reaction time t0 and the safe lane-changing distance d0; the lane-changing ending point p e is a point on the lane center line of the ending lane.

[0007] calling a preset lane-changing path simulator to simulate a driving path from the lane-changing starting point p s to the lane-changing ending point p e and generate a corresponding first simulation path;

[0008] loading each of a plurality of path planning models to be evaluated, and calling the currently loaded path planning model to plan a driving path from the lane-changing starting point p s to the lane-changing ending point p e and generate a corresponding first planning path;

[0009] evaluating the path similarity of the first simulation path and each of the first planning paths to obtain corresponding first and second similarities;

[0010] composing corresponding first model evaluation data from the model name of each of the path planning models and the corresponding first and second similarities, and composing a corresponding first evaluation report from all the obtained first model evaluation data.

[0011] Preferably, the corresponding lane-changing ending point p e is determined according to the lane-changing starting point p s , the longitudinal vehicle speed v0, the driver reaction time t0 and the safe lane-changing distance d0, and specifically includes:

[0012] calculating a lane-changing distance d1 according to the longitudinal vehicle speed v0, the driver reaction time t0 and the safe lane-changing distance d0, d1 = v0*t0+d0; and taking a point on the lane center line of the ending lane with a longitudinal distance of the lane-changing distance d1 from the lane-changing starting point p s as the corresponding lane-changing ending point p e .

[0013] Preferably, the calling of the preset lane-changing path simulator to simulate a driving path from the lane-changing starting point p s to the lane-changing ending point p e and generate a corresponding first simulation path specifically includes:

[0014] the lane-changing path simulator performing Bezier curve characteristic point setting according to the lane-changing distance d1, the lane-changing starting point p s and the lane-changing ending point p e to obtain corresponding first, second, third and fourth characteristic points p1, p2, p3 and p4;

[0015] According to the first, second, third and fourth feature points p1, p2, p3 and p4, a third-order Bezier curve is planned to obtain the corresponding first simulation path.

[0016] Further, the lane-changing path simulator sets the first, second, third and fourth feature points p1, p2, p3 and p4 according to the lane-changing distance d1, the lane-changing starting point p s and the lane-changing ending point p e , and specifically includes:

[0017] The lane-changing path simulator calculates a corresponding feature point distance d2 according to a preset proportion parameter k and the lane-changing distance d1, d2=k*d1.

[0018] The lane-changing starting point p s is taken as the corresponding first feature point p1.

[0019] Starting from the lane-changing starting point p s , a tangent line is drawn along the lane center line of the starting lane in the vehicle driving direction to obtain a corresponding first tangent line; and a point on the first tangent line with a longitudinal distance of the feature point distance d2 from the lane-changing starting point p s is taken as the corresponding second feature point p2.

[0020] Starting from the lane-changing ending point p e , a tangent line is drawn along the lane center line of the ending lane in the opposite direction of the vehicle driving direction to obtain a corresponding second tangent line; and a point on the second tangent line with a longitudinal distance of the feature point distance d2 from the lane-changing ending point p e is taken as the corresponding third feature point p3.

[0021] The lane-changing ending point p e is taken as the corresponding fourth feature point p4.

[0022] Further, the lane-changing path simulator plans a third-order Bezier curve according to the first, second, third and fourth feature points p1, p2, p3 and p4 to obtain the corresponding first simulation path, and specifically includes:

[0023] The first, second, third and fourth feature points p1, p2, p3 and p4 are brought into a third-order Bezier curve equation to obtain a plurality of sampling points b i .

[0024] b i =(1-r i ) 3 ·p1+3r i ·(1-r i ) 2• p2+3r i 2 • (1-r i )·p3+r i 3 • p4;

[0025] wherein, i is a sampling point index, i≥0; r i is an equation proportionality parameter, the equation proportionality parameter r i is a ratio of a longitudinal distance between the sampling point b i and the first feature point p1 and a longitudinal distance between the first feature point p1 and the fourth feature point p4, 0≤r i ≤1, r i -r i-1 =△d / d1,△d is a preset longitudinal unit distance;

[0026] According to the longitudinal unit distance△d and the longitudinal vehicle speed v0, the corresponding sampling point time t i of each sampling point b i is set, t i＝0 =0, t i =t i-1 +△d / v0;

[0027] The corresponding simulation path point is composed of each sampling point b i and the corresponding sampling point time t i , and the corresponding simulation path point sequence is composed of all the simulation path points obtained; and the corresponding first simulation path is generated by curve fitting according to the simulation path point sequence.

[0028] Preferably, the path similarity of the first simulation path and each first planning path is evaluated to obtain the corresponding first similarity and second similarity, specifically including:

[0029] The first simulation path and the current first planning path are respectively sampled at the same time sampling interval to obtain the corresponding first simulation path point sequence and first planning path point sequence; and the number of first simulation path points A j in the first simulation path point sequence is counted to generate the corresponding sampling point number N; the first simulation path point sequence includes a plurality of first simulation path points A j ; the first planning path point sequence includes a plurality of first planning path points B j ; 1≤j≤N; the first simulation path point A j corresponds to the first planning path point B j one by one according to the sampling point index j;

[0030] The lane change starting point ps A two-dimensional Cartesian coordinate system is constructed, taking the origin, the tangent of the lane center line of the starting lane as the Y axis, and the driving direction of the vehicle as the positive direction of the Y axis, denoted as a first coordinate system; and the first simulation path points A j , the first planning path points B j in the first coordinate system are denoted as corresponding first projection points A and second projection points B

[0031] In the first coordinate system, the Euclidean distances of the first simulation path point sequence and the first planning path point sequence are calculated to obtain corresponding first Euclidean distances D s , and the first Euclidean distances D s are taken as the first similarity corresponding to the current first planning path;

[0032] In the first coordinate system, the vectors from the origin to the first projection points A and the second projection points B are denoted as corresponding first vectors A and second vectors B and the vector included angle cosines of the first vectors A and the corresponding second vectors B are calculated to generate corresponding first included angle cosines C j , and the sum of all the first included angle cosines C j is calculated to generate a corresponding first included angle cosine sum C s , and the first included angle cosine sum C s is taken as the second similarity corresponding to the current first planning path.

[0033] The second aspect of the embodiment of the application provides an electronic device, comprising a memory, a processor and a transceiver;

[0034] The processor is used for coupling with the memory, reading and executing instructions in the memory, so as to realize the method steps of the first aspect;

[0035] The transceiver is coupled with the processor, and the transceiver is controlled by the processor to perform message transmission and reception.

[0036] The third aspect of the embodiment of the application provides a computer readable storage medium, which stores computer instructions, when the computer instructions are executed by a computer, the computer instructions make the computer execute the instructions of the method of the first aspect.

[0037] The embodiment of the present application provides a processing method for evaluating a lane-changing path by using a simulation platform, an electronic device and a computer readable storage medium; a simulation platform is used to evaluate and provide a lane-changing path simulator; the simulation platform creates and sets an evaluation road scene before evaluation; the simulation platform calls the lane-changing path simulator to plan a path under the current evaluation road scene to obtain a corresponding simulation path, then calls each path planning model to be evaluated to plan a path to obtain a corresponding planning path, and then evaluates the curve spacing and curve shape of each planning path with reference to the simulation path to obtain corresponding first and second similarities (the smaller the first and second similarities are, the more similar they are), and finally generates an evaluation report according to the obtained first and second similarities. The present application provides a general scheme for evaluating the performance of multiple alternative path planning models, improves the selection ability of an automatic driving system for path planning models, improves the path planning performance of the automatic driving system, and improves the guarantee ability of the automatic driving system for safe driving. BRIEF DESCRIPTION OF DRAWINGS

[0038] Figure 1 A processing method for evaluating a lane-changing path by using a simulation platform provided by the embodiment of the present application is shown in the schematic diagram.

[0039] Figure 2a An evaluation road scene provided by the embodiment of the present application is shown in the schematic diagram.

[0040] Figure 2b A Bezier curve feature point provided by the embodiment of the present application is shown in the schematic diagram.

[0041] Figure 3 A structure schematic diagram of an electronic device provided by the embodiment of the present application is shown in the schematic diagram. DETAILED DESCRIPTION

[0042] In order to make the purpose, technical scheme and advantages of the present application clearer, the present application will be further described in detail below with reference to the drawings. Obviously, the described embodiments are only part of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0043] The embodiment of the present application provides a processing method for evaluating a lane-changing path by using a simulation platform, an electronic device and a computer readable storage medium; a simulation platform is used to evaluate and provide a lane-changing path simulator; the simulation platform creates and sets an evaluation road scene before evaluation; the simulation platform calls the lane-changing path simulator to plan a path under the current evaluation road scene to obtain a corresponding simulation path, then calls each path planning model to be evaluated to plan a path to obtain a corresponding planning path, and then evaluates the curve spacing and curve shape of each planning path with reference to the simulation path to obtain corresponding first and second similarities (the smaller the first and second similarities are, the more similar they are), and finally generates an evaluation report according to the obtained first and second similarities. The present application provides a general scheme for evaluating the performance of multiple alternative path planning models, improves the selection ability of an automatic driving system for path planning models, improves the path planning performance of the automatic driving system, and improves the guarantee ability of the automatic driving system for safe driving. Figure 1 A processing method for evaluating a lane-changing path by using a simulation platform provided by the embodiment of the present application is shown in the schematic diagram.

[0044] Step 1, the simulation platform initializes an evaluation road scene;

[0045] The road scenario being evaluated includes the starting lane and the ending lane.

[0046] Here, the simulation platform is specifically a simulation platform for an autonomous driving system. The simulation platform for an autonomous driving system creates and initializes an evaluation road scenario by calling road test or simulation data. In this embodiment of the invention, the evaluation road scenario is a scenario consisting of multiple parallel lanes, where two lanes are respectively set as the starting lane and the ending lane. The starting lane is the starting lane for the current lane change simulation, and the ending lane is the target lane for the current lane change simulation.

[0047] Step 2: Set the starting point of the center line of the starting lane as the lane change starting point p. s The longitudinal vehicle speed v0, driver reaction time t0, and safe lane change distance d0 are set; and the lane change starting point p is used as the basis for setting these parameters. s The longitudinal vehicle speed v0, driver reaction time t0, and safe lane change distance d0 determine the corresponding lane change end point p. e ;

[0048] Among them, the lane change end point p e This refers to a point on the center line of the lane that ends;

[0049] Specifically, this includes: Step 21, setting the starting point of the centerline of the starting lane's centerline as the lane change starting point p. s And set the longitudinal vehicle speed v0, driver reaction time t0 and safe lane change distance d0;

[0050] Here, after the simulation platform completes the initialization of the road scenario for evaluation, it also needs to set some system parameters for lane change simulation in the scenario. These system parameters include at least the lane change starting point p. s The parameters used for lane change simulation include longitudinal vehicle speed v0, driver reaction time t0, and safe lane change distance d0; while another system parameter used for lane change simulation is the lane change end point p. e Then, estimation is required through subsequent steps;

[0051] Step 22, based on the lane change starting point p s The longitudinal vehicle speed v0, driver reaction time t0, and safe lane change distance d0 determine the corresponding lane change end point p. e ;

[0052] Specifically, this includes: calculating the lane change distance d1 based on the longitudinal vehicle speed v0, driver reaction time t0, and safe lane change distance d0, where d1 = v0 * t0 + d0; and aligning the center line of the ending lane with the lane change starting point p. s The point whose longitudinal distance is the lane change distance d1 is taken as the corresponding lane change end point p. e .

[0053] Here, with Figure 2a Taking the schematic diagram of the evaluation road scenario provided in Embodiment 1 of the present invention as an example, the various parameters in the evaluation road scenario are explained.

[0054] Step 3: Call the preset lane change path simulator to perform a lane change from the starting point p. s To the end point of the lane change p e The driving path is simulated to generate the corresponding first simulated path;

[0055] Here, the lane change path simulator in this embodiment of the invention is a curve simulator based on the third-order Bézier curve equation. The first simulated path obtained by the simulation platform calling the lane change path simulator to perform ideal path curve simulation is the reference curve used for subsequent evaluation.

[0056] Specifically, this includes: Step 31, the lane change path simulator calculates the lane change distance d1 and the lane change starting point p. s and lane change end point p e By setting the feature points of the Bézier curve, the corresponding first, second, third, and fourth feature points p1, p2, p3, and p4 are obtained;

[0057] Specifically, it includes: Step 311, the lane change path simulator calculates the corresponding feature point distance d2 according to the preset ratio parameter k and the lane change distance d1, d2=k*d1;

[0058] Here, the scaling parameter k is a pre-set empirical parameter, which is set to 0.35 by default;

[0059] Step 312, change the lane change starting point p s As the corresponding first feature point p1;

[0060] Step 313, from the lane change starting point p s Starting from the beginning, draw a tangent line along the vehicle's direction of travel to the center line of the starting lane to obtain the first tangent line; then connect this first tangent line to the lane change starting point p. s The point whose vertical distance is the feature point distance d2 is taken as the corresponding second feature point p2;

[0061] Step 314, from the lane change end point p e Starting from the beginning, draw a tangent to the center line of the ending lane in the opposite direction of vehicle travel to obtain the corresponding second tangent; then connect the second tangent to the lane change end point p. e The point whose vertical distance is the feature point distance d2 is taken as the corresponding third feature point p3;

[0062] Step 315, change the lane change end point p e p4 is the corresponding fourth feature point;

[0063] Here, withFigure 2b Taking the schematic diagram of Bézier curve feature points provided in Embodiment 1 of the present invention as an example, the first, second, third, and fourth feature points p1, p2, p3, and p4 will be described.

[0064] Step 32: Based on the first, second, third, and fourth feature points p1, p2, p3, and p4, perform third-order Bézier curve planning to obtain the corresponding first simulated path;

[0065] Specifically, this includes step 321, substituting the first, second, third, and fourth feature points p1, p2, p3, and p4 into the third-order Bézier curve equation to calculate multiple sampling points b. i :

[0066] b i =(1-r i ) 3 ·p1+3r i ·(1-r i ) 2 ·p2+3r i 2 ·(1-r i )·p3+r i 3 p4;

[0067] Where i is the sampling point index, i≥0; r i Let r be the scaling parameter of the equation. i For the corresponding sampling point b i The ratio of the vertical distance between the first feature point p1 and the vertical distance between the first feature point p1 and the fourth feature point p4, 0 ≤ r i ≤1, r i -r i-1 = △d / d1, where △d is the preset vertical unit spacing;

[0068] Here, the above equation is the equation of a third-order Bézier curve, given the proportionality parameter r of the equation. i Given four feature points p1, p2, p3, and p4, the position of each sampling point, i.e., sampling point b, can be determined. i It should be noted that, in this embodiment of the invention, the original coordinates of the first and second feature points p1 and p2 are defaulted to the SL coordinate system coordinates of the starting lane, and the original coordinates of the third and fourth feature points p3 and p4 are defaulted to the SL coordinate system coordinates of the ending lane. Before substituting the first, second, third, and fourth feature points p1, p2, p3, and p4 into the third-order Bézier curve equation, this embodiment of the invention will pre-calculate the lane change starting point p. sA two-dimensional Cartesian coordinate system is constructed with the origin as the origin, the tangent to the center line of the starting lane as the Y-axis, and the vehicle's direction of travel as the positive Y-axis. This system is denoted as the current two-dimensional coordinate system. The first, second, third, and fourth feature points p1, p2, p3, and p4 are projected onto the current two-dimensional coordinate system to obtain the coordinates of four corresponding projected points in the current two-dimensional coordinate system. Then, these four corresponding projected point coordinates are substituted into the third-order Bézier curve equation to calculate multiple sampling points b in the current two-dimensional coordinate system. i ;

[0069] Step 322, based on the longitudinal unit spacing Δd and the longitudinal vehicle speed v0, sample points b are... i The corresponding sampling time t i Configure, t i＝0 =0,t i =t i-1 +△d / v0;

[0070] Here, the longitudinal unit spacing Δd is a pre-set empirical parameter; furthermore, this embodiment of the invention also stipulates that the longitudinal driving speed of the lane change path simulator and the multiple path planning models in subsequent steps remains constant during path planning; based on this longitudinal constant speed mode, when the longitudinal unit spacing Δd is known, the longitudinal speed of each sampling point b can be determined. i The time point information is the sampling time t. i To confirm;

[0071] Step 323, from each sampling point b i and the corresponding sampling time t i The corresponding simulated path points are formed, and the obtained simulated path points form the corresponding simulated path point sequence; and the first simulated path is generated by curve fitting based on the simulated path point sequence.

[0072] Here, the first simulated path is the ideal path curve simulated by the lane change path simulator, which will be used as the reference curve for evaluation in subsequent steps.

[0073] Step 4: Load each of the multiple path planning models to be evaluated; and call the currently loaded path planning model to evaluate the path from the lane change starting point p. s To the end point of the lane change p e The driving route is planned to generate the corresponding first planned route.

[0074] Here, each path planning model is based on the same scenario parameters as the lane change path simulator (lane change starting point p). s Longitudinal vehicle speed v0, driver reaction time t0, safe lane change distance d0, lane change end point p eThe same longitudinal constant speed mode is used to perform path planning to obtain their respective planned path curves, i.e., the first planned path.

[0075] Step 5: Evaluate the path similarity between the first simulated path and each of the first planned paths to obtain the corresponding first similarity and second similarity.

[0076] Here, the simulation platform of this invention uses the simulated path as a reference to evaluate the curve spacing and curve shape of each planned path to obtain the corresponding first and second similarities. The smaller the first similarity, the closer the two compared paths are; the smaller the second similarity, the more similar the shapes of the two compared paths are.

[0077] Specifically, this includes: Step 51, sampling the first simulated path and the current first planned path at the same time sampling interval to obtain the corresponding first simulated path point sequence and first planned path point sequence; and sampling the first simulated path point A of the first simulated path point sequence. j The number of samples is counted to generate the corresponding number of sampling points N;

[0078] The first simulated path point sequence includes multiple first simulated path points A. j The first planned path point sequence includes multiple first planned path points B. j ; 1≤j≤N; First simulated path point A j With the first planned path point B j Each sampling point corresponds one-to-one with the sampling point index j;

[0079] Step 52, starting from lane change point p s Using the origin as the origin, a two-dimensional Cartesian coordinate system is constructed with the tangent of the center line of the starting lane as the Y-axis and the vehicle's direction of travel as the positive Y-axis, denoted as the first coordinate system; and each of the first simulated path points A... j First planned path point B j The projection point in the first coordinate system is denoted as the corresponding first projection point. Second projection point

[0080] Here, because the coordinate system of the first planned path output by each path planning model may be different, it is necessary to unify the first planned path and the first simulated path under the same coordinate system; as mentioned above, the coordinate system of the first simulated path is based on the lane change starting point p. s A two-dimensional Cartesian coordinate system is constructed with the origin as the origin, the tangent to the center line of the starting lane as the Y-axis, and the vehicle's direction of travel as the positive Y-axis. Therefore, the same two-dimensional Cartesian coordinate system, namely the first coordinate system, must be created and the first planned path and the first simulated path must be unified under this coordinate system. Here, since the first simulated path itself is based on the first coordinate system, each point A of the first simulated path...j With the corresponding first projection point The coordinates have not changed;

[0081] Step 53: In the first coordinate system, calculate the corresponding first Euclidean distance D between the first simulated path point sequence and the first planned path point sequence. s , And the first Euclidean distance D s As the first similarity to the current first planning path;

[0082] here, The Euclidean distance is the distance between two matching path pairs on the first simulated path and the first planned path that occur at the same time; the first Euclidean distance D s The first similarity is the sum of the Euclidean distances between all matching path point pairs in the first simulated path and the first planned path. The smaller the first similarity, the smaller the distance between the first simulated path and the first planned path, and the closer the two paths are. Conversely, the larger the first similarity, the larger the distance between the first simulated path and the first planned path, and the farther apart the two paths are.

[0083] Step 54: In the first coordinate system, transfer the origin to each first projection point. and each of the second projection points The vector is denoted as the corresponding first vector. Second vector And for each first vector and the corresponding second vector The first cosine C is generated by calculating the cosine of the included angle between the vectors. j , And the cosines of all the first included angles C obtained j Perform a summation calculation to generate the corresponding sum of the first angle cosines C. s , The sum of the cosines of the first included angle C s This serves as the second similarity to the current first planning path.

[0084] Here, the cosine of the first included angle C j Let C be the cosine of the orientation angle between two matching path point pairs with the same time on the first simulated path and the first planned path. j The smaller the value, the smaller the orientation difference between the matched path point pairs; conversely, the larger the value, the greater the difference. The sum of the cosines of the first included angle, C... sThe second similarity is the sum of the cosine values ​​of the orientation angles of all matching path point pairs in the first simulated path and the first planned path; the smaller the second similarity, the smaller the shape difference between the first simulated path and the first planned path, and the more similar their shapes are. Conversely, the larger the second similarity, the more obvious the shape difference between the first simulated path and the first planned path, and the less similar their shapes are.

[0085] Step 6: The first model evaluation data is composed of the model name of each path planning model and the corresponding first similarity and second similarity; and the first evaluation report is composed of all the obtained first model evaluation data.

[0086] Here, the simulation platform of this embodiment of the invention repeats step 5 to obtain the first and second similarities of each path planning model. Then, the evaluation data of each path planning model, i.e., the first model evaluation data, is obtained by combining the model name of the path planning model with the first similarity and the second similarity. By analyzing the data in the first evaluation report provided by the simulation platform, the staff or the autonomous driving system can confirm the performance level of the multiple path planning models evaluated in this instance.

[0087] Figure 3 This is a schematic diagram of an electronic device provided in Embodiment 2 of the present invention. This electronic device can be the aforementioned terminal device or server, or it can be a terminal device or server connected to the aforementioned terminal device or server that implements the method of the embodiments of the present invention. Figure 3 As shown, the electronic device may include: a processor 301 (e.g., CPU), a memory 302, and a transceiver 303; the transceiver 303 is coupled to the processor 301, and the processor 301 controls the transmission and reception operations of the transceiver 303. The memory 302 may store various instructions for performing various processing functions and implementing the processing steps described in the foregoing method embodiments. Preferably, the electronic device involved in the embodiments of the present invention further includes: a power supply 304, a system bus 305, and a communication port 306. The system bus 305 is used to realize communication connections between components. The communication port 306 is used for communication between the electronic device and other peripherals.

[0088] exist Figure 3The system bus 305 mentioned can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. This system bus can be divided into address bus, data bus, control bus, etc. For ease of representation, it is represented by only one thick line in the figure, but this does not indicate that there is only one bus or one type of bus. The communication interface is used to enable communication between the database access device and other devices (e.g., clients, read-write libraries, and read-only libraries). Memory may include Random Access Memory (RAM) and may also include non-volatile memory, such as at least one disk storage device.

[0089] The processors mentioned above can be general-purpose processors, including central processing units (CPUs), network processors (NPs), graphics processing units (GPUs), etc.; they can also be digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.

[0090] It should be noted that the embodiments of the present invention also provide a computer-readable storage medium storing instructions that, when run on a computer, cause the computer to perform the methods and processes provided in the above embodiments.

[0091] This invention also provides a chip for executing instructions, which is used to perform the processing steps described in the foregoing method embodiments.

[0092] This invention provides a method, electronic device, and computer-readable storage medium for evaluating lane change paths using a simulation platform. The method utilizes the simulation platform's capabilities for evaluation and provides a lane change path simulator. Before evaluation, the simulation platform creates and sets up the evaluation road scenario. During evaluation, the platform calls the lane change path simulator to plan the corresponding simulated path in the current evaluation road scenario. Then, it calls each path planning model to be evaluated to plan the corresponding planned path. Using the simulated path as a reference, it evaluates the curve spacing and curve shape of each planned path to obtain corresponding first and second similarities (the smaller the first and second similarities, the more similar the paths). Finally, it generates a corresponding evaluation report based on the obtained first and second similarities. This invention provides a general solution for evaluating the performance of multiple alternative path planning models, improving the autonomous driving system's ability to select path planning models, enhancing the path planning performance of the autonomous driving system, and improving the autonomous driving system's ability to ensure safe driving.

[0093] Those skilled in the art will further recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

[0094] The steps of the methods or algorithms described in conjunction with the embodiments disclosed herein can be implemented in hardware, a software module executed by a processor, or a combination of both. The software module can be located in random access memory (RAM), main memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, removable disk, CD-ROM, or any other form of storage medium known in the art.

[0095] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for evaluating lane change paths using a simulation platform, characterized in that, The method includes: The simulation platform initializes and evaluates a road scenario; the evaluated road scenario includes a starting lane and an ending lane. Set the starting point of the centerline of the starting lane's centerline as the lane change starting point p. s The longitudinal vehicle speed v0, driver reaction time t0, and safe lane change distance d0 are set; and the lane change starting point p is used as the reference. s The longitudinal vehicle speed v0, the driver's reaction time t0, and the safe lane change distance d0 determine the corresponding lane change end point p. e The lane change end point p e Let be a point on the center line of the lane where the lane ends; Call the preset lane change path simulator to perform the lane change from the starting point p s To the lane change end point p e The driving path is simulated to generate the corresponding first simulated path; Load each of the multiple path planning models to be evaluated one by one; and call the currently loaded path planning model to evaluate the path from the lane change starting point p. s To the lane change end point p e The driving route is planned to generate the corresponding first planned route; The path similarity between the first simulated path and each of the first planned paths is evaluated to obtain the corresponding first similarity and second similarity. The first model evaluation data consists of the model name of each path planning model and the corresponding first similarity and second similarity; and the first evaluation report consists of all the obtained first model evaluation data. Specifically, the evaluation of the path similarity between the first simulated path and each of the first planned paths to obtain the corresponding first similarity and second similarity includes: The first simulated path and the current first planned path are sampled at the same time sampling interval to obtain the corresponding first simulated path point sequence and first planned path point sequence; and the first simulated path point A of the first simulated path point sequence is then sampled. j The number of samples is statistically analyzed to generate the corresponding number of sampling points N; the first simulated path point sequence includes multiple first simulated path points A. j The first planned path point sequence includes multiple first planned path points B. j ; 1≤j≤N; the first simulated path point A j With the first planned path point B j Each sampling point corresponds one-to-one with the sampling point index j; Using the lane change starting point p s Using the origin as the origin, a two-dimensional Cartesian coordinate system is constructed with the tangent of the center line of the starting lane as the Y-axis and the vehicle's direction of travel as the positive Y-axis, denoted as the first coordinate system; and each of the first simulated path points A... j The first planned path point B j The projection point in the first coordinate system is denoted as the corresponding first projection point. Second projection point ; In the first coordinate system, the Euclidean distance D is calculated between the first simulated path point sequence and the first planned path point sequence. s , ; and the first Euclidean distance D s As the first similarity to the current first planned path; In the first coordinate system, the origin is mapped to each of the first projection points. and each of the second projection points The vector is denoted as the corresponding first vector. Second vector ; and for each of the first vectors and the corresponding second vector The first cosine C is generated by calculating the cosine of the included angle between the vectors. j , ; and for all the cosines C of the first included angle obtained j Perform a summation calculation to generate the corresponding sum of the first angle cosines C. s , ; and the sum of the cosines of the first included angle C s This serves as the second similarity to the current first planned path.

2. The processing method for evaluating lane change paths using a simulation platform according to claim 1, characterized in that, The lane change starting point p s The longitudinal vehicle speed v0, the driver's reaction time t0, and the safe lane change distance d0 determine the corresponding lane change end point p. e Specifically, it includes: The lane change distance d1 is calculated based on the longitudinal vehicle speed v0, the driver's reaction time t0, and the safe lane change distance d0, where d1 = v0 * t0 + d0; and the center line of the ending lane is aligned with the lane change starting point p. s The point whose longitudinal distance is the lane change distance d1 is taken as the corresponding lane change end point p. e .

3. The method for evaluating lane change paths using a simulation platform according to claim 2, characterized in that, The preset lane change path simulator is invoked to perform a lane change from the lane change starting point p. s To the lane change end point p e The driving path is simulated to generate the corresponding first simulated path, specifically including: The lane change path simulator is based on the lane change distance d1 and the lane change starting point p. s and the lane change end point p e By setting the feature points of the Bézier curve, the corresponding first, second, third, and fourth feature points p1, p2, p3, and p4 are obtained; The first simulated path is obtained by performing third-order Bézier curve planning based on the first, second, third, and fourth feature points p1, p2, p3, and p4.

4. The processing method for evaluating lane change paths using a simulation platform according to claim 3, characterized in that, The lane change path simulator is based on the lane change distance d1 and the lane change starting point p. s and the lane change end point p e By setting the feature points of the Bézier curve, the corresponding first, second, third, and fourth feature points p1, p2, p3, and p4 are obtained, specifically including: The lane change path simulator calculates the corresponding feature point distance d2 based on the preset ratio parameter k and the lane change distance d1, where d2 = k * d1. The lane change starting point p s As the corresponding first feature point p1; From the lane change starting point p s Starting from the beginning, draw a tangent line along the vehicle's direction of travel to the center line of the starting lane to obtain the corresponding first tangent line; and connect the first tangent line with the lane change starting point p. s The point whose vertical distance is the distance d2 between the feature points is taken as the corresponding second feature point p2; From the lane change end point p e Starting from the beginning, draw a tangent to the center line of the lane ending in the opposite direction of vehicle travel to obtain a corresponding second tangent; and connect the second tangent to the lane change end point p. e The point whose vertical distance is the distance d2 of the feature point is taken as the corresponding third feature point p3; The lane change end point p e This corresponds to the fourth feature point p4.

5. The processing method for evaluating lane change paths using a simulation platform according to claim 3, characterized in that, The step of obtaining the first simulated path by performing third-order Bézier curve planning based on the first, second, third, and fourth feature points p1, p2, p3, and p4 specifically includes: Substituting the first, second, third, and fourth feature points p1, p2, p3, and p4 into the third-order Bézier curve equation, multiple sampling points b are obtained. i : ； Where i is the sampling point index, i≥0; r i The equation's scaling parameter r is the scaling parameter of the equation. i For the corresponding sampling point b i The ratio of the vertical distance between the first feature point p1 and the vertical distance between the first feature point p1 and the fourth feature point p4, 0 ≤ r i ≤1, r i -r i-1 =△d / d1, where △d is the preset vertical unit spacing; Based on the longitudinal unit spacing Δd and the longitudinal vehicle speed v0, each of the sampling points b i The corresponding sampling time t i Configure, t i=0 =0,t i =t i-1 +△d / v0; From each of the sampling points b i and the corresponding sampling time t i The corresponding simulated path points are formed, and all the obtained simulated path points form a corresponding simulated path point sequence; and the corresponding first simulated path is generated by curve fitting based on the simulated path point sequence.

6. An electronic device, characterized in that, include: Memory, processor, and transceiver; The processor is configured to be coupled to the memory, read and execute instructions in the memory to implement the method according to any one of claims 1-5; The transceiver is coupled to the processor, and the processor controls the transceiver to send and receive messages.

7. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that, when executed by a computer, cause the computer to perform the method described in any one of claims 1-5.

Images