A guiding wire closed-loop simulation verification method and system

Abstract

The present invention provides a method and system for closed-loop simulation verification of a guiding line, which are used to verify the reliability and comfort of the guiding line. The guiding line generated by the guiding line generation module is given to the simulation system. The simulation system verifies the guiding line according to the difference between the simulated driving trajectory and the guiding line trajectory, and then feeds back the verification result to the guiding line generation module. The guiding line generation module generates a more reliable and comfortable guiding line based on the simulation result, and finally outputs it to the simulation system for verification, thus forming a simulation system with a data closed-loop until the generated guiding line is reliable and comfortable. The present invention has low cost and is easy to implement. It does not require a real vehicle to participate. Only a simulation scenario needs to be built, and the verification result can be obtained by simulating driving in the simulation environment.

Description

Technical Field

[0001] The present invention relates to the field of high-precision map crowdsourcing composition and simulation, and more specifically, to a method and system for closed-loop simulation verification of guiding lines. Background Art

[0002] Due to its advantages such as low cost, wide data collection range, and real-time update, crowdsourcing has become one of the mainstream methods for constructing high-precision maps. A guiding line is a recommended route that conforms to driving habits generated based on crowdsourcing trajectories and is used to guide intelligent driving vehicles to drive safely and smoothly in a barrier-free environment. In scenarios where lane boundaries can be sensed, the center line of the left and right sensed boundaries of the lane is usually directly selected as the lane guiding line. In scenarios where there are no sensed boundaries, such as at intersections, a passable trajectory needs to be generated through curve fitting or by mining historical trajectory data within the intersection as the guiding line within the intersection.

[0003] As a key element of the crowdsourcing map, the verification of its usability is particularly important. Real vehicle verification is extremely costly, and currently, the conditions for the actual implementation of autonomous driving are limited, making it difficult to achieve full-map coverage verification of guiding lines. Summary of the Invention

[0004] The present invention provides a method and system for closed-loop simulation verification of guiding lines to address the technical problems existing in the prior art.

[0005] According to a first aspect of the present invention, a method for closed-loop simulation verification of guiding lines is provided, including:

[0006] S1, building a simulation environment and configuring a vehicle dynamics model;

[0007] S2, controlling the vehicle to perform path tracking along the guiding line trajectory in the simulation environment according to the generated guiding line trajectory;

[0008] S3, calculating the lateral difference and lateral force coefficient distribution between the output simulation trajectory and the guiding line trajectory to verify the reliability and comfort of the guiding line trajectory;

[0009] S4, if the guiding line trajectory is unreliable or uncomfortable, adjusting the generated guiding line trajectory based on the simulation trajectory, and executing S2 to S4 until the generated guiding line trajectory is reliable and comfortable.

[0010] Based on the above technical solutions, the present invention can also be improved as follows.

[0011] Optionally, building the simulation environment in S1 includes:

[0012] Build a road network in the Unreal Engine based on crowdsourced map data, and create a road surface physical material. Set the friction in the material parameters to 1, the friction merge mode to "multiply", the restitution to 0, and set the physical material parameters in the road surface material to the just-created physical material;

[0013] The configured vehicle dynamics model includes:

[0014] Enable the physical simulation of the skeletal mesh: Select the skeletal mesh component for physical simulation, set the vehicle steering mode to front-wheel steering, and set the front-wheel angle and wheel damping ratio of the vehicle.

[0015] Optionally, before S2, it also includes:

[0016] Based on the historical trajectories in the extracted crowdsourced map, use the GAN network deep learning model as the guiding line generation module to generate guiding line trajectories.

[0017] Optionally, S2 includes:

[0018] According to the current state of the vehicle and the target trajectory, based on the MPC control algorithm, output the steering wheel control amount to perform lateral control on the vehicle, and keep the lateral control error within the set distance range;

[0019] According to the current state of the vehicle and the target trajectory, based on the PID control algorithm, output the throttle and brake control amounts to perform longitudinal control on the vehicle, and keep the speed error within the set speed range, where the guiding line trajectory is the target trajectory.

[0020] Optionally, S3 includes:

[0021] Divide the current guiding line trajectory into Sn segments of guiding line sub-trajectories according to the set length distance, and divide the simulation trajectory into Sn segments of simulation sub-trajectories according to the set length distance;

[0022] According to each segment of the guiding line sub-trajectory and each segment of the simulation sub-trajectory, judge whether the deviation of each segment of the simulation sub-trajectory is abnormal, and count the first quantity ratio of the simulation sub-trajectories with abnormal deviations to the simulation sub-trajectories;

[0023] When the first quantity ratio is greater than the first set ratio, determine that the current guiding line trajectory is unreliable, otherwise, determine that the current guiding line trajectory is reliable;

[0024] And,

[0025] According to each segment of the guiding line sub-trajectory and each segment of the simulation sub-trajectory, judge whether the lateral force coefficient of each segment of the simulation sub-trajectory is abnormal, and count the second quantity ratio of the simulation sub-trajectories with abnormal lateral force coefficients to the simulation trajectory;

[0026] If the second quantity ratio is greater than the second set ratio, it is determined that the current guide line trajectory is uncomfortable; otherwise, it is determined that the current guide line trajectory is comfortable.

[0027] Optionally, based on each segment of the guide line sub-trajectory and each segment of the simulation sub-trajectory, determining whether the deviation of each segment of the simulation sub-trajectory is abnormal, and counting the first quantity ratio of the simulation sub-trajectories with abnormal deviations to the simulation sub-trajectories, includes:

[0028] During the simulation run, the current position P0 of the vehicle is obtained every set length distance, and the point P1 closest to the current position P0 of the vehicle is found on the current guide line, and the distance between P0 and P1 is calculated. If the distance is greater than the set distance threshold, the deviation of the corresponding simulation sub-trajectory is abnormal;

[0029] Count the number S0 of simulation sub-trajectories with abnormal deviations, and calculate the first quantity ratio PCT0 = S0 / S n .

[0030] Optionally, based on each segment of the guide line sub-trajectory and each segment of the simulation sub-trajectory, determining whether the lateral force coefficient of each segment of the simulation sub-trajectory is abnormal, and counting the second quantity ratio of the simulation sub-trajectories with abnormal lateral force coefficients to the simulation trajectory, includes:

[0031] During the simulation run, the lateral force coefficient of the vehicle is calculated every set length distance, where the lateral force coefficient is the ratio of the lateral acceleration to the gravitational acceleration;

[0032] If the lateral force coefficient is greater than the set coefficient threshold, the lateral force coefficient of the corresponding simulation sub-trajectory is abnormal;

[0033] Count the number S1 of simulation sub-trajectories with abnormal lateral force coefficients, and calculate the second quantity ratio PCT2 = S1 / S n .

[0034] Optionally, after S3, it further includes performing a secondary simulation to determine the unreliable reasons for the unreliable guide line, including:

[0035] For the unreliable guide line, set the target speed of the vehicle, and perform a second trace driving and reliability evaluation on the guide line;

[0036] If the result of the secondary evaluation is reliable, the unreliable reason for the guide line is that the speed of the guide line is too high; otherwise, the unreliable reason for the guide line is that the turning radius of the guide line is too small;

[0037] Output the guide line simulation result, and the simulation result includes the guide line ID, the first quantity ratio PCT0, the second quantity ratio PCT1, the reliability situation, the comfort situation, the unreliable reason, and the simulation trajectory shape point string.

[0038] Optionally, in S4, if the guide line trajectory is unreliable or uncomfortable, based on the simulation trajectory to adjust the generated guide line trajectory, execute S2 to S4 until the generated guide line trajectory is reliable and comfortable, including:

[0039] The guide line generation module adjusts the speed or shape of the guide line trajectory according to the unreliable reason, regenerates the guide line trajectory, and executes S2 to S4 until the generated guide line trajectory is reliable and comfortable.

[0040] According to a second aspect of the present invention, there is provided a guide line closed-loop simulation verification system, including:

[0041] A building module, configured to build a simulation environment and configure a vehicle dynamics model;

[0042] A control module, configured to control the vehicle to perform path tracking driving along the guide line trajectory in the simulation environment according to the generated guide line trajectory;

[0043] A verification module, configured to calculate the lateral difference and the lateral force coefficient distribution between the output simulation trajectory and the guide line trajectory to verify the reliability and comfort of the guide line trajectory;

[0044] An adjustment module, configured to, if the guide line trajectory is unreliable or uncomfortable, adjust the generated guide line trajectory based on the simulation trajectory, and repeatedly call the control module, the verification module, and the adjustment module until the generated guide line trajectory is reliable and comfortable.

[0045] According to a third aspect of the present invention, there is provided an electronic device, including a memory and a processor, and the processor is configured to implement the steps of the guide line closed-loop simulation verification method when executing a computer management program stored in the memory.

[0046] According to a fourth aspect of the present invention, there is provided a computer-readable storage medium, on which a computer management program is stored, and the computer management program is configured to implement the steps of the guide line closed-loop simulation verification method when executed by a processor.

[0047] A guide line closed-loop simulation verification method and system provided by the present invention send the guide line generated by the guide line generation module to the simulation system. The simulation system verifies the guide line according to the difference between the simulation driving trajectory and the guide line trajectory, and then feeds back the verification result to the guide line generation module. The guide line generation module generates a more reliable and comfortable guide line through the simulation result, and finally outputs it to the simulation system for verification, thereby forming a data-closed-loop simulation system until the generated guide line is reliable and comfortable. The present invention has low cost and is easy to implement, does not require a real vehicle to participate, only needs to build a simulation scenario, and can obtain the verification result by simulating driving in the simulation environment. Description of the Drawings

[0048] Figure 1 Flow chart of a method for closed-loop simulation verification of a guiding line provided by the present invention;

[0049] Figure 2 Schematic diagram of verification indexes for the reliability and comfort of the guiding line;

[0050] Figure 3 Flow chart of specific implementation steps for verifying the reliability and comfort of each guiding line;

[0051] Figure 4 Overall flow chart of a method for closed-loop simulation verification of a guiding line;

[0052] Figure 5 Schematic diagram of the structure of a system for closed-loop simulation verification of a guiding line provided by the present invention;

[0053] Figure 6 Schematic diagram of the hardware structure of a possible electronic device provided by the present invention;

[0054] Figure 7 Schematic diagram of the hardware structure of a possible computer-readable storage medium provided by the present invention. Specific embodiments

[0055] To make the objectives, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention. In addition, the technical features in each embodiment or a single embodiment provided by the present invention can be combined with each other arbitrarily to form a feasible technical solution. This kind of combination is not restricted by the order of steps and / or the mode of structural composition, but must be based on what can be achieved by those of ordinary skill in the art. When the combination of technical solutions appears to be contradictory or unable to be implemented, it should be considered that such a combination of technical solutions does not exist and is not within the protection scope required by the present invention.

[0056] The present invention uses crowdsourced map data to construct a simulation road network in the Unreal Engine, verifies the reliability of the guiding line by comparing the differences between the simulated driving trajectory and the guiding line trajectory, and verifies the comfort of the guiding line by the lateral force coefficient of the vehicle during operation.

[0057] Among them, the input of the simulation system is the crowdsourced guiding line, and then the vehicle dynamics model is controlled by the MPC and PID control algorithms to perform path tracking driving along the guiding line, and the reliability and comfort of the guiding line are judged by the vehicle state during operation.

[0058] Figure 1 The flowchart of a closed-loop simulation verification method for a guiding line provided by the present invention is as Figure 1 shown, and the method includes:

[0059] S1. Build a simulation environment and configure a vehicle dynamics model.

[0060] It can be understood that to ensure the credibility of the simulation results, the present invention uses the open-source Unreal Engine as the simulation platform, and the vehicle dynamics model adopts the PhyX Vehicle physics engine. As one of the three major physics engines, the PhyX engine has been widely used in major simulation platforms.

[0061] First, build a simulation scene: use crowdsourced map data to build a road network in the Unreal Engine, and create the physical material of the road surface. In the material parameters, the friction is set to 1, the friction combination mode is set to "multiply", the restitution force is set to 0, and the physical material parameters in the road surface material are set to the just-created physical material.

[0062] After building the simulation environment, configure the vehicle dynamics model: enable the physical simulation of the skeletal mesh. The method is to select the skeletal mesh component -> Physics -> Simulate Physics and check. Set the vehicle steering mode to front-wheel steering, and the front-wheel steering angle is set to 40°. The steering angle of a general car is 35° to 45°, so the present system takes 40°. The wheel damping ratio is set to 0.3, and the maximum friction force obtained by multiplying with the road surface friction is 0.3, which is close to the maximum friction coefficient of the real cement road surface.

[0063] As an embodiment, before the said S2, it further includes: based on the historical trajectories extracted from the crowdsourced map, using the GAN network deep learning model as a guiding line generation module to generate guiding line trajectories.

[0064] It can be understood that the mainstream GAN network deep learning model is used as the guiding line generation module, and by extracting the historical trajectories from the crowdsourced map, the guiding line is generated as the input of the simulation system.

[0065] S2. According to the generated guiding line trajectory, control the vehicle to perform path tracking driving along the guiding line trajectory in the simulation environment.

[0066] As an embodiment, the said S2 includes: based on the current state of the vehicle and the target trajectory, output the steering wheel control quantity based on the MPC control algorithm to perform lateral control on the vehicle, and keep the lateral control error within the set distance range; based on the current state of the vehicle and the target trajectory, output the throttle and brake control quantities based on the PID control algorithm to perform longitudinal control on the vehicle, and keep the speed error within the set speed range, where the guiding line trajectory is the target trajectory.

[0067] It is understandable that the vehicle lateral control adopts the MPC control algorithm, inputs the current state and target trajectory of the vehicle, and outputs the steering wheel control amount; the longitudinal control adopts the PID control, inputs the current state and target trajectory of the vehicle, and outputs the throttle and brake control amounts. To ensure the reliability of the simulation results, it is required that the lateral control error is within 0.3 meters and the speed error is controlled within 1 m / s. That is to say, when driving along a straight line or a large-radius curve, the lateral distance between the vehicle and the guiding line must be controlled within 0.3 meters, and the error between the current speed of the vehicle and the target speed is within 1 m / s. Otherwise, the reliability of the simulation results will be affected.

[0068] S3. Calculate the lateral difference and lateral force coefficient distribution between the output simulation trajectory and the guiding line trajectory to verify the reliability and comfort of the guiding line trajectory.

[0069] It is understandable that the S3 includes: dividing the current guiding line trajectory into Sn segments of guiding line sub-trajectories according to a set length distance, and dividing the simulation trajectory into Sn segments of simulation sub-trajectories according to the set length distance; judging whether the deviation of each segment of the simulation sub-trajectory is abnormal according to each segment of the guiding line sub-trajectory and each segment of the simulation sub-trajectory, and counting the first quantity ratio of the simulation sub-trajectories with abnormal deviations to the simulation sub-trajectories; when the first quantity ratio is greater than the first set ratio, it is determined that the current guiding line trajectory is unreliable, otherwise, it is determined that the current guiding line trajectory is reliable.

[0070] For each guiding line, perform the following operations:

[0071] Divide the current guiding line trajectory into Sn segments of guiding line sub-trajectories according to a set length distance, and divide the simulation trajectory into Sn segments of simulation sub-trajectories according to the set length distance. Among them, the set length distance can be set to 20 ms. See Figure 2 and Figure 3 , and perform reliability evaluation and comfort evaluation on the guiding line trajectory respectively.

[0072] (1) Verify the reliability of the guiding line.

[0073] When the distance between the vehicle and the guiding line during path following driving exceeds 0.5 m, it is recorded as an abnormal deviation, and the statistical count of the abnormal deviation is set as S0. During the simulation run, the current position P0 of the vehicle is obtained every 20 ms. The simulation trajectory within these 20 ms is called a simulation sub-trajectory. The point P1 on the current guiding line that is closest to the vehicle is found, and the distance between P0 and P1 is calculated. If the distance is greater than 0.5 m, then S0 is incremented by 1, that is, the deviation of this section of the simulation sub-trajectory is abnormal. After a section of the guiding line reaches the end, the percentage PCT0 of the abnormal deviation is calculated, that is, the ratio of S0 to Sn, which is called the first quantity ratio. This first quantity ratio is used as an evaluation index for the reliability of the guiding line. When PCT0 > 0.25, that is, more than 25% of the simulation trajectory points have a lateral gap with the guiding line exceeding 0.5 m, then the guiding line is considered unreliable; otherwise, the guiding line is considered reliable.

[0074] As an embodiment, according to each section of the guiding line sub-trajectory and each section of the simulation sub-trajectory, it is determined whether the lateral force coefficient of each section of the simulation sub-trajectory is abnormal, and the second quantity ratio of the simulation sub-trajectories with abnormal lateral force coefficients to the simulation trajectory is statistically counted; when the second quantity ratio is greater than the second set ratio, it is determined that the current guiding line trajectory is uncomfortable; otherwise, it is determined that the current guiding line trajectory is comfortable.

[0075] (2) Conduct comfort verification.

[0076] When the lateral force coefficient of the vehicle during path following driving exceeds 0.35, it is recorded as an abnormal lateral force coefficient, and the statistical count of the abnormal lateral force coefficient is set as S1. During the operation, the lateral force coefficient of the vehicle is calculated every 20 ms. If the lateral force coefficient is greater than 0.35, then S1 is incremented by 1. The calculation method of the lateral force coefficient is the ratio of the lateral acceleration to the gravitational acceleration. The lateral acceleration can be directly obtained in the simulation, and the gravitational acceleration is taken as 9.8. Among them, the table of the influence of the lateral force coefficient on the passenger comfort is as follows in Table 1.

[0077] Table 1 Source basis table of the evaluation index of the lateral force coefficient

[0078]

[0079]

[0080] After a section of the guiding line reaches the end, the percentage PCT1 of the abnormal lateral force coefficient is calculated, that is, the ratio of S1 to Sn, which is called the second quantity ratio. This second quantity ratio is used as an evaluation index for the comfort of the guiding line. When PCT1 > 0.20, that is, more than 20% of the simulation trajectory points have a lateral force coefficient exceeding 0.35, then the guiding line is considered uncomfortable; otherwise, the guiding line is considered comfortable.

[0081] S4. If the generated guide line trajectory is unreliable or uncomfortable, adjust the generated guide line trajectory based on the simulation trajectory, and execute S2 to S4 until the generated guide line trajectory is reliable and comfortable.

[0082] As an embodiment, after S3, secondary simulation is further included to determine the reasons for the unreliability of the unreliable guide line, including: for the unreliable guide line, set the target speed of the vehicle, perform the second trace driving and reliability evaluation on the guide line; if the secondary evaluation result is reliable, the reason for the unreliability of the guide line is that the speed of the guide line is too high, otherwise, the reason for the unreliability of the guide line is that the turning radius of the guide line is too small; output the guide line simulation result, and the simulation result includes the guide line ID, the first quantity ratio PCT0, the second quantity ratio PCT1, the reliability situation, the comfort situation, the reason for unreliability, and the simulation trajectory shape point string.

[0083] It can be understood that for the unreliable guide line, the reasons for the unreliability of the guide line are determined through secondary simulation. Specifically, for the guide line with the verification result being unreliable, set the target speed of the vehicle to 10 km / h, perform the second trace driving and reliability evaluation on this section of the guide line. If the secondary evaluation result is reliable, it proves that the reason for the unreliability of this section of the guide line is that the speed of the guide line is too high, otherwise, it indicates that it is caused by too small a turning radius, and the guide line generation module needs to adjust the speed or shape of the guide line according to the reason.

[0084] Finally, output the guide line simulation result, and the simulation result includes the guide line ID, the abnormal deviation ratio PCT0, the abnormal lateral force coefficient ratio PCT1, the reliability situation, the comfort situation, the reason for unreliability, and the simulation trajectory shape point string.

[0085] For the guide line with the simulation result being unreliable or uncomfortable, feedback the simulation result to the guide line generation module, and repeatedly execute steps 2, 3, 4, and 5 until the guide line is reliable and comfortable. The above process can form a data-closed-loop simulation-based guide line generation method.

[0086] See Figure 4 , which is the overall flowchart of a guide line closed-loop simulation verification method. The input of this simulation method is the crowdsourced guide line, and then the vehicle dynamics model is controlled by the MPC and PID control algorithms to perform trace driving along the guide line, and the reliability and comfort of the guide line are judged through the vehicle state during the running process. The steps are as follows:

[0087] Step 1: Build a simulation environment and configure the vehicle dynamics model. Generate a simulation road network through the crowdsourced map in the Unreal Engine as the driving environment of the vehicle, create a vehicle model, and configure the model parameters. The closer the simulation environment parameters are to the reality, the more realistic and reliable the simulation result will be.

[0088] Step 2: Generate a guiding line. Use a mainstream GAN network deep learning model as the guiding line generation module. By extracting historical trajectories from crowd-sourced maps, generate a guiding line as the input to the simulation system.

[0089] Step 3: Trace following driving. Through MPC and PID control algorithms, control the vehicle to perform trace following driving along the guiding line. The control algorithm should make the driving trajectory and speed of the vehicle as consistent as possible with the input guiding line to ensure the reliability of the simulation results.

[0090] Step 4: Verify the reliability and comfort of the guiding line through the lateral difference between the simulation trajectory and the guiding line trajectory and the distribution of lateral force coefficients.

[0091] Step 5: For guiding lines with unreliable or uncomfortable simulation results, feed back the simulation trajectory and results to the guiding line generation module, and repeatedly execute Steps 2, 3, 4, and 5 until all guiding lines are reliable and comfortable.

[0092] See Figure 5 , a guiding line closed-loop simulation verification system provided by the present invention. The system includes a construction module 501, a control module 502, a verification module 503, and an adjustment module 504, where:

[0093] The construction module 501 is used to construct a simulation environment and configure a vehicle dynamics model;

[0094] The control module 502 is used to control the vehicle to perform trace following driving along the guiding line trajectory in the simulation environment according to the generated guiding line trajectory;

[0095] The verification module 503 is used to calculate the lateral difference between the output simulation trajectory and the guiding line trajectory and the distribution of lateral force coefficients to verify the reliability and comfort of the guiding line trajectory;

[0096] The adjustment module 504 is used to, if the guiding line trajectory is unreliable or uncomfortable, adjust the generated guiding line trajectory based on the simulation trajectory, and repeatedly call the control module, the verification module, and the adjustment module until the generated guiding line trajectory is reliable and comfortable.

[0097] It can be understood that a guiding line closed-loop simulation verification system provided by the present invention corresponds to the guiding line closed-loop simulation verification method provided in the foregoing embodiments. The relevant technical features of the guiding line closed-loop simulation verification system can refer to the relevant technical features of the guiding line closed-loop simulation verification method, which will not be elaborated herein.

[0098] Please refer to Figure 6 , Figure 6 is a schematic diagram of an embodiment of an electronic device provided by an embodiment of the present invention. As Figure 6As shown in the figure, an embodiment of the present invention provides an electronic device 600, including a memory 610, a processor 620, and a computer program 611 stored on the memory 610 and executable on the processor 620. When the processor 620 executes the computer program 611, the steps of the guiding line closed-loop simulation verification method are implemented.

[0099] Please refer to Figure 7 , Figure 7 which is a schematic diagram of an embodiment of a computer-readable storage medium provided by the present invention. As Figure 7 shown, this embodiment provides a computer-readable storage medium 700, on which a computer program 711 is stored. When the computer program 711 is executed by a processor, the steps of the guiding line closed-loop simulation verification method are implemented.

[0100] A guiding line closed-loop simulation verification method and system provided by an embodiment of the present invention have the following advantages:

[0101] (1) Low cost and easy to implement: It does not require a real vehicle to participate. Only a simulation scenario needs to be built, and driving can be simulated in the simulation environment to obtain the verification result.

[0102] (2) It can provide a reference trajectory for the guiding line generation module: The driving trajectory output by the simulation conforms to vehicle dynamics and can be used as a reference trajectory for the guiding line generation module to generate a more reliable and comfortable guiding line in the crowdsourcing map.

[0103] It should be noted that in the above embodiments, the descriptions of the various embodiments have their own focuses. For the parts not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0104] Those skilled in the art should understand that the embodiments of the present invention can be provided as a method, a system, or a computer program product. Therefore, the present invention can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk memories, CD-ROMs, optical memories, etc.) containing computer-usable program codes.

[0105] The present invention is described with reference to the flowcharts and / or block diagrams of methods, apparatuses (systems), and computer program products according to embodiments of the present invention. It should be understood that each flow and / or block in the flowchart and / or block diagram, and the combinations of flows and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded computer, or other programmable data processing devices to generate a machine, such that the instructions executed by the processor of the computer or other programmable data processing devices generate means for implementing the functions specified in one flow Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0106] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, such that the instructions stored in the computer-readable memory generate a manufactured article including instruction means that implement the functions specified in one flow Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0107] These computer program instructions can also be loaded onto a computer or other programmable data processing device, such that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process, so that the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in one flow Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0108] Although the preferred embodiments of the present invention have been described, those skilled in the art can make additional changes and modifications to these embodiments once they know the basic creative concepts. Therefore, the appended claims are intended to be construed to include the preferred embodiments as well as all changes and modifications that fall within the scope of the present invention.

[0109] Obviously, those skilled in the art can make various changes and variations to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalent technologies, the present invention is also intended to include these modifications and variations.

Claims

1. A method for closed-loop simulation verification of a guiding wire, characterized in that, Including: S1. Build a simulation environment and configure a vehicle dynamics model; S2. According to the generated guiding line trajectory, control the vehicle to perform path tracking driving along the guiding line trajectory in the simulation environment; S3. Calculate the lateral difference and lateral force coefficient distribution between the output simulation trajectory and the guiding line trajectory to verify the reliability and comfort of the guiding line trajectory; S4. If the guiding line trajectory is unreliable or uncomfortable, adjust the generated guiding line trajectory based on the simulation trajectory, and execute S2 to S4 until the generated guiding line trajectory is reliable and comfortable; The S3 includes: Divide the current guiding line trajectory into Sn segments of guiding line sub-trajectories according to a set length distance, and divide the simulation trajectory into Sn segments of simulation sub-trajectories according to the set length distance; According to each segment of guiding line sub-trajectory and each segment of simulation sub-trajectory, judge whether the deviation of each segment of simulation sub-trajectory is abnormal, and count the first quantity proportion of the simulation sub-trajectories with abnormal deviations among the simulation sub-trajectories; When the first quantity proportion is greater than the first set proportion, determine that the current guiding line trajectory is unreliable, otherwise, determine that the current guiding line trajectory is reliable; And, According to each segment of guiding line sub-trajectory and each segment of simulation sub-trajectory, judge whether the lateral force coefficient of each segment of simulation sub-trajectory is abnormal, and count the second quantity proportion of the simulation sub-trajectories with abnormal lateral force coefficients among the simulation trajectories; When the second quantity proportion is greater than the second set proportion, determine that the current guiding line trajectory is uncomfortable, otherwise, determine that the current guiding line trajectory is comfortable.

2. The guiding line closed-loop simulation verification method according to claim 1, characterized in that Building the simulation environment in S1 includes: Build a road network in the Unreal Engine based on crowdsourced map data, and create a road surface physical material. Set the friction in the material parameters to 1, the friction merge mode to "multiply", the restoring force to 0, and set the physical material parameters in the road surface material to the just-created physical material; The configuring the vehicle dynamics model includes: Enable the physical simulation of the skeletal mesh: Select the skeletal mesh component for physical simulation, set the vehicle steering mode to front-wheel steering, and set the front-wheel angle and wheel damping ratio of the vehicle.

3. The guiding line closed-loop simulation verification method according to claim 1, characterized in that Before S2, it also includes: Based on the historical trajectories extracted from the crowdsourced map, use the GAN network deep learning model as a guiding line generation module to generate a guiding line trajectory.

4. The guiding wire closed-loop simulation verification method according to claim 1, characterized in that, The S2 includes: According to the current state of the vehicle and the target trajectory, output a steering wheel control amount based on the MPC control algorithm to perform lateral control on the vehicle, and keep its lateral control error within a set distance range; According to the current state of the vehicle and the target trajectory, output throttle and brake control amounts based on the PID control algorithm to perform longitudinal control on the vehicle, and keep its speed error within a set speed range, where the guiding line trajectory is the target trajectory.

5. The guiding wire closed-loop simulation verification method according to claim 1, wherein The judging whether the deviation of each segment of simulation sub-trajectory is abnormal according to each segment of guiding line sub-trajectory and each segment of simulation sub-trajectory, and counting the first quantity proportion of the simulation sub-trajectories with abnormal deviations among the simulation sub-trajectories includes: During the simulation run, the current position P0 of the vehicle is obtained at intervals of a set length distance, and the point P1 closest to the current position P0 of the vehicle is found on the current guiding line. The distance between P0 and P1 is calculated. If the distance is greater than the set distance threshold, the deviation of the corresponding simulation sub-trajectory is abnormal; Statistically count the number S0 of abnormal simulation sub-trajectories with deviation, and calculate the first quantity proportion PCT0 = S0 / S n .

6. The guiding wire closed-loop simulation verification method according to claim 1, characterized in that Judging whether the lateral force coefficient of each simulation sub-trajectory is abnormal according to each guiding line sub-trajectory and each simulation sub-trajectory, and counting the second quantity ratio of the simulation sub-trajectories with abnormal lateral force coefficients to the simulation trajectory, including: During the simulation run, the lateral force coefficient of the vehicle is calculated at intervals of a set length distance, where the lateral force coefficient is the ratio of the lateral acceleration to the gravitational acceleration; If the lateral force coefficient is greater than the set coefficient threshold, the lateral force coefficient of the corresponding simulation sub-trajectory is abnormal; Statistically count the number S1 of abnormal simulation sub-trajectories of the transverse force coefficient, and calculate the second quantity ratio PCT2 = S1 / S n .

7. The guiding wire closed-loop simulation verification method according to claim 1, wherein After S3, it also includes performing a secondary simulation to determine the reason for the unreliability of the unreliable guiding line, including: For the unreliable guiding line, set the target speed of the vehicle, and perform a second trace driving and reliability evaluation on the guiding line; If the result of the secondary evaluation is reliable, the reason for the unreliability of the guiding line is that the speed of the guiding line is too high, otherwise, the reason for the unreliability of the guiding line is that the turning radius of the guiding line is too small; Output the guiding line simulation result, where the simulation result includes the guiding line ID, the first quantity ratio PCT0, the second quantity ratio PCT1, the reliability situation, the comfort situation, the reason for unreliability, and the simulation trajectory shape point string.

8. The lead wire closed-loop simulation verification method according to claim 7, characterized in that In S4, if the guiding line trajectory is unreliable or uncomfortable, based on the simulation trajectory to adjust the generated guiding line trajectory, execute S2 to S4 until the generated guiding line trajectory is reliable and comfortable, including: The guiding line generation module adjusts the speed or shape of the guiding line trajectory according to the reason for unreliability, regenerates the guiding line trajectory, and executes S2 to S4 until the generated guiding line trajectory is reliable and comfortable.

9. A guiding wire closed-loop simulation verification system, characterized in that, Including: A building module for building a simulation environment and configuring a vehicle dynamics model; A control module for controlling the vehicle to perform trace driving along the guiding line trajectory in the simulation environment according to the generated guiding line trajectory; A verification module for calculating the lateral difference and lateral force coefficient distribution between the output simulation trajectory and the guiding line trajectory to verify the reliability and comfort of the guiding line trajectory; An adjustment module for, if the guiding line trajectory is unreliable or uncomfortable, adjusting the generated guiding line trajectory based on the simulation trajectory, and repeatedly calling the control module, the verification module, and the adjustment module until the generated guiding line trajectory is reliable and comfortable; The verification module is specifically used for: Dividing the current guiding line trajectory into Sn guiding line sub-trajectories according to the set length distance, and dividing the simulation trajectory into Sn simulation sub-trajectories according to the set length distance; According to each guiding line sub-trajectory and each simulation sub-trajectory, judging whether the deviation of each simulation sub-trajectory is abnormal, and counting the first quantity ratio of the simulation sub-trajectories with abnormal deviations to the simulation sub-trajectories; When the first quantity ratio is greater than the first set ratio, it is determined that the current guiding line trajectory is unreliable, otherwise, it is determined that the current guiding line trajectory is reliable; And, Based on each segment of the guiding line sub-trajectory and each segment of the simulation sub-trajectory, determine whether the lateral force coefficient of each segment of the simulation sub-trajectory is abnormal, and count the proportion of the second quantity of the simulation sub-trajectories with abnormal lateral force coefficients in the simulation trajectory; When the proportion of the second quantity is greater than the second set ratio, it is determined that the current guiding line trajectory is uncomfortable; otherwise, it is determined that the current guiding line trajectory is comfortable.

Images