System for generating multi-strategy trajectories in high-precision maps based on pid control

Abstract

The application belongs to the technical field of automatic driving simulation test, and particularly relates to a system for generating multiple strategy trajectories in a high-precision map based on PID control.The system comprises a vehicle control module, a lane control module, a global control module and a high-precision map module, wherein the lane control module is connected with the vehicle control module, the global control module is connected with the lane control module, and the high-precision map module is connected with the global control module; the PID control algorithm and heuristic rules can be used to generate multiple strategy trajectories of a vehicle in an automatic driving simulation scene, including acceleration, deceleration, straight driving and left-right lane changing strategies.The application can generate multiple strategy trajectories, which are close to real scenes.

Description

Technical Field

[0001] This invention belongs to the field of autonomous driving simulation testing technology, specifically involving a system for generating multi-strategy trajectories using high-precision map information and PID control algorithms. Background Technology

[0002] Simulation testing is an essential part of autonomous driving technology. Its purpose is to use a virtual environment to simulate a real environment and verify the algorithm performance of autonomous vehicles. In simulation testing, the virtual environment not only needs to provide accurate map information, but also simulated information about traffic participants, including vehicles and pedestrians. These traffic participants need to travel along a certain route and interact with each other to make reasonable decisions, so as to provide an autonomous vehicle with a more realistic simulation environment. Summary of the Invention

[0003] The purpose of this invention is to provide a system based on PID control that can efficiently generate multi-strategy trajectories for all vehicles in a traffic scenario using multiple strategies in a high-precision map.

[0004] This invention provides a system for generating multi-strategy trajectories in high-precision maps based on PID control. According to a physical model and heuristic rules, it generates trajectories under multiple strategies, closely resembling real-world scenarios. Specifically, it includes a vehicle control module, a lane control module, a global control module, and a high-precision map module. The global control module is connected to several lane control modules, the lane control modules are connected to several vehicle control modules, and the high-precision map module is connected to the global control module. All of these modules are deployed within a computer and can be accessed by simulation testing software.

[0005] The vehicle control module is used to control the trajectory generation of a vehicle, including maintaining the vehicle's current basic state and information about its lane (coordinates, speed, orientation angle, steering wheel angle, target speed, lane and strategy information), obtaining vehicle information in front of and behind the vehicle in its lane and vehicle information in adjacent lanes from the lane control module, and generating the vehicle's future state and trajectory based on the obtained information and PID control method.

[0006] The lane control module is used to control the vehicle control module corresponding to the vehicles in a lane, maintain the front and rear sequence information of the vehicles in the lane, provide information interaction between different vehicles to the controlled vehicle control module, and support the interaction between the controlled vehicle control module and the global control module.

[0007] The global control module is used to control all lane control modules, obtain the coordinate information of the lane centerline from the high-precision map module, maintain the current basic status and strategy information of all vehicles, and record the generated trajectory of all vehicles.

[0008] The high-precision map module is used to import high-precision map information, store lane centerline information in the map information using a data structure, support the location query of vehicle coordinates in the lane, and provide the point sequence representation of the lane centerline and the information of adjacent lanes.

[0009] Furthermore, the vehicle control module specifically includes:

[0010] The vehicle speed PID control unit is used to obtain the vehicle's current speed v in time frame t. t and the target rate of the current time frame t Calculate the current rate s t =|v t |, based on the difference between the current rate and the target rate The PID algorithm is used to calculate the rate compensation value, thereby calculating the rate s of the next time frame. t+1 In the PID algorithm, the coefficients of the proportional, integral, and derivative terms are adjustable control parameters, denoted as S. p ,S i ,S d ;

[0011] The vehicle steering wheel angle PID control unit is used to obtain the vehicle's current speed v at time frame t. t Position coordinates p t Orientation o t Steering wheel angle a t The following lane center line c t Based on the following information, the future position coordinates of the vehicle five time frames after the current state are calculated using the model below:

[0012]

[0013] Where Δt is the time interval between adjacent time frames, we can set Δt = 0.1 seconds, p' t+5 For the predicted future location coordinates after 5 time frames, o t The vehicle's orientation (i.e., angle, starting from the positive x-axis in a Cartesian coordinate system and increasing clockwise), v t p is the vehicle's current speed. t This is the vehicle's current location;

[0014] Then, the current position p of the vehicle is calculated using the following formula. t Distance c from the center line of the lane being followed t The nearest point position p c :

[0015]

[0016] The lane centerline consists of a sequence of m two-dimensional point vectors, denoted as . d is the distance p from the center line of the lane. t The endpoint number of the nearest line segment; DIS(X,Y,Z) represents the distance from point X to the line segment formed by connecting points Y and Z; CP(X,Y,Z) represents the distance from point X to the nearest point on the line segment formed by connecting points Y and Z; then, the following formula is used to calculate the distance from point p. c Starting from the current location, travel the corresponding distance along the center line of the lane to reach the future target coordinates.

[0017]

[0018] The lane centerline consists of a sequence of m two-dimensional point vectors, denoted as . Δt is the time interval between adjacent frames, v t d represents the vehicle's current speed, l represents the distance the vehicle will travel in the next 5 time frames, and d represents the lane centerline. t The distance from the vehicle's current position p t The endpoint number of the nearest line segment, L i Indicates from p t Start driving along the center line of the lane to the destination. The length of the distance traveled; e represents L d ,L d+1 …L m-1 The largest subscript whose length does not exceed l; Indicates from p c Starting from the current position, travel a distance *l* along the lane centerline to reach the future target coordinates. Then, calculate the ratio of the distance between the future position coordinates and the future target coordinates to the current speed. As a result of the discrepancy, a PID algorithm is used to calculate the compensation value for the steering wheel angle, thereby calculating the steering wheel angle 'a' for the next time frame. t+1 In the PID algorithm, the coefficients of the proportional, integral, and derivative terms are adjustable control parameters, denoted as A. p A i A d .

[0019] The strategy trajectory generation unit obtains the position and speed information of vehicles surrounding the current vehicle from the lane control module, and selects from strategies such as acceleration, deceleration, maintaining straightness, and changing lanes left or right for the current vehicle. It adjusts the target speed and target lane centerline of the vehicle speed PID control unit and the vehicle steering wheel angle PID control unit to generate the corresponding strategy trajectory; specifically:

[0020] For acceleration / deceleration strategies, the current target rate is adjusted by the vehicle speed PID control unit. To achieve this, using The target rate is replaced by Δs, where Δs is the target rate increment. When Δs is positive, it indicates an acceleration strategy; when Δs is negative, it indicates a deceleration strategy.

[0021] For the straight-ahead strategy, no adjustment is made; for the left / right lane-changing strategy, the current lane centerline c of the vehicle's steering wheel angle PID control unit is adjusted. t To achieve this, use C. t left or right lane center line c' t+1 To replace the center line of the following lane;

[0022] The current strategy for selecting a vehicle is determined by heuristic rules based on the vehicle's driving status and information about surrounding vehicles. When the speed of the vehicle in front is less than the current speed of the vehicle and there is a risk of rear-end collision, the strategy of deceleration or lane change is selected; when the speed of the vehicle in front is greater than the current speed of the vehicle, the strategy of acceleration or maintaining straight driving is selected; if there are multiple feasible strategies in the current state, a random selection strategy is introduced.

[0023] Furthermore, the lane control module includes:

[0024] The vehicle management unit is used to manage the vehicle control modules corresponding to vehicles in the lane, maintain the front and rear sequence information of vehicles, and support operations such as adding, deleting, and querying vehicle information in the lane.

[0025] The inter-lane information interaction unit is used to support information interaction between different lane control modules. By maintaining a linked list structure, it maintains the numbers of the left and right adjacent lanes for each lane, thereby supporting vehicle information interaction between adjacent lanes and assisting the lane control module in generating lane change strategy trajectories.

[0026] Furthermore, the global control module includes:

[0027] The lane management unit is used to manage all lane control modules, supports the creation, deletion, and querying of lane control modules, and interacts with all vehicle control modules through the lane control modules.

[0028] The trajectory generation unit is used to invoke all vehicle control modules through the lane management unit. When generating a trajectory, it first selects a strategy for all vehicle control modules, and then invokes each vehicle control module sequentially. In each invocation, each vehicle control module calculates the vehicle state for the next time frame using a physical model based on the information of the vehicle it controls. The specific formula is as follows:

[0029]

[0030] Where Δt represents the time interval between adjacent time frames, pt Indicates the vehicle's current position, v t Indicates the vehicle's current speed, o t Indicates the vehicle's current orientation angle, a t The current steering wheel angle of the vehicle is indicated by l, which represents the distance between the front and rear axles. t+1 Indicates the position of the vehicle in the next time frame, o t+1 This indicates the vehicle's orientation in the next time frame; subsequently, the above process is iterated N times to generate a strategy trajectory for N consecutive time frames. The trajectory generation unit will then use the trajectory information p t+1 ,p t+2 …,p t+N Stored in the computer's storage medium.

[0031] Furthermore, the high-precision map module includes:

[0032] The map information storage unit includes a storage medium in a computer for storing map information, including the coordinate information of lane center lines in the map. Any lane center line is composed of a two-dimensional point coordinate sequence, which is represented as a polyline formed by connecting the points in the sequence in sequence.

[0033] The map information query unit, implemented using a gridded data structure, supports efficient query operations on map information, including querying the coordinates of the center line of a specified lane, querying the information of adjacent lanes, and querying the center line of the lane closest to the coordinates of a given vehicle.

[0034] The system for generating multi-strategy trajectories in high-precision maps based on PID control has the following specific workflow:

[0035] (1) Obtain the vehicle's initialization information as system input, including the vehicle's coordinates, speed, orientation angle, steering wheel angle, target speed, and strategy information;

[0036] (2) Use the high-precision map module to import map information, including the coordinate information of the lane center line, and locate the lane of each vehicle;

[0037] (3) Use the global control module to create lane control modules for all lanes, and use the lane control modules to create vehicle control modules for each vehicle and manage them, and generate the strategy trajectory for the next time frame for each vehicle.

[0038] (4) Use the global control module to iterate multiple times to generate a strategy trajectory for each vehicle for a continuous period of time.

[0039] The main technical features and functional advantages of this invention are as follows:

[0040] By using PID control algorithms and heuristic rules, various strategy trajectories of vehicles in autonomous driving simulation scenarios are generated, including strategies such as acceleration, deceleration, maintaining straight-line driving, and changing lanes left or right. Attached Figure Description

[0041] Figure 1 This is a structural diagram of the system for generating multi-strategy trajectories in high-precision maps based on PID control, as proposed in this invention.

[0042] Figure 2 This is a schematic diagram of an implementation scenario of the method for generating multi-strategy trajectories in high-precision maps based on PID control proposed in this invention.

[0043] Figure 3 This is the simulation experiment scenario for the present invention. Detailed Implementation

[0044] The present invention will be further described below with reference to the embodiments and accompanying drawings.

[0045] Figure 1 This is a schematic diagram of the system for generating multi-strategy trajectories in a high-precision map based on PID control, as proposed in this invention. The system consists of four main modules: a vehicle control module, a lane control module, a global control module, and a high-precision map module. The lane control module is connected to the vehicle control module; the global control module is connected to the lane control module; and the high-precision map module is connected to the global control module.

[0046] Figure 2 This is a schematic diagram of an implementation scenario of the method for generating multi-strategy trajectories in high-precision maps based on PID control proposed in this invention.

[0047] In this embodiment, there are three vehicles: 201, 202, and 203. Vehicle 202 is set as an obstacle with a speed of 0 in the scene, and no trajectory is generated for it. The dot pointed to by 204 represents the strategy trajectories generated for vehicles 201 and 203 in this invention: for vehicle 201, there is an obstacle blocking its lane, so a right lane change strategy trajectory is generated; for vehicle 203, there is no obstacle blocking its lane, so a straight acceleration strategy trajectory is generated for it; the diamond-shaped dot pointed to by 205 and its connecting line represent the center lines of the two upward-moving lanes, left and right. Vehicle 201 originally followed the center line of the left lane, but during the generation of the lane change strategy trajectory, vehicle 201 changed to follow the center line of the right lane.

[0048] The vehicle control module includes a vehicle speed PID control unit, a vehicle steering wheel angle PID control unit, and a strategy trajectory generation unit.

[0049] The vehicle speed PID control unit is used to obtain the vehicle's current speed v in time frame t. t and the target rate of the current time frame t Where v t Given a two-dimensional vector, the target speed It is a scalar; further, the current rate s is calculated. t =|v t |, based on the difference between the current rate and the target rate The PID algorithm is used to calculate the rate compensation value, and the current speed is added to the compensation value to calculate the rate s of the next time frame. t+1 The PID algorithm uses S p =0.1,S i =0,S d =0.01 as the PID control parameter;

[0050] The vehicle steering wheel angle PID control unit is used to obtain the vehicle's current speed v at time frame t. t Position coordinates p t Orientation o t Steering wheel angle a t The following lane center line c t Information such as p t For two-dimensional coordinates, o t and a t All are in radians, c t The line represents a sequence of two-dimensional point coordinates; next, the future position coordinates of the vehicle five time frames after the current state are calculated based on the following model:

[0051]

[0052] Where Δt = 0.1 seconds is the time interval between adjacent time frames, and p' t+5 For the predicted future location coordinates after 5 time frames, o t For the vehicle's orientation, v t p is the vehicle's current speed. t Let p be the vehicle's current position; then, computational geometry algorithms are used to calculate the vehicle's current position p. t Distance c from the center line of the lane being followed t The position of the nearest point p c Then, computational geometry algorithms are used to calculate from p c Starting from the position, proceed along the center line of the lane for 5Δt·|v t |Distance to the future target coordinates Next, the ratio of the distance between the future position coordinates and the future target coordinates to the current velocity magnitude is calculated. As a difference, this ratio is used to reflect the difference between the current motion state of the vehicle and the motion state of the following lane, so as to use the PID algorithm to calculate the compensation value of the steering wheel angle.

[0053] Furthermore, by adding the current steering wheel angle to the steering wheel angle compensation value, the steering wheel angle 'a' for the next time frame is calculated. t+1 The PID algorithm uses A p =0.5,A i =0,A d =0.02 as the PID control parameter;

[0054] The strategy trajectory generation unit is used to obtain the position and speed information of vehicles around the current vehicle from the lane control module, and select from strategies such as acceleration, deceleration, maintaining straightness, and changing lanes left or right for the current vehicle. It adjusts the target speed and target lane centerline of the vehicle speed PID control unit and the vehicle steering wheel angle PID control unit to generate a corresponding strategy trajectory; specifically... Figure 2 In the scenario shown, vehicle 201 obtains the position and speed information of vehicle 202 ahead from the lane control module, determines that there is an obstacle, needs to decelerate or change lanes, and selects a lane change strategy. Since there is no drivable lane on the left in the current lane, the strategy of changing lanes to the right is selected; vehicle 203 obtains from the lane control module that there are no other vehicles ahead, so the strategy of keeping straight is selected.

[0055] Furthermore, the lane control module includes a vehicle management unit and an inter-lane information interaction unit;

[0056] The vehicle management unit is used to manage the vehicle control modules corresponding to vehicles in the lane, maintain the front and rear sequence information of vehicles, and support operations such as adding, deleting, and querying vehicle information in the lane.

[0057] The inter-lane information interaction unit is used to support information interaction between different lane control modules. By maintaining a linked list structure, it maintains the numbers of the left and right adjacent lanes for each lane, thereby supporting vehicle information interaction between adjacent lanes and assisting the lane control module in generating lane change strategy trajectories.

[0058] Furthermore, the global control module includes a lane management unit and a trajectory generation unit;

[0059] The lane management unit is used to manage all lane control modules, supports the creation, deletion and query operations of lane control modules, and interacts with all vehicle control modules through the lane control modules;

[0060] The trajectory generation unit is used to call upon all vehicle control modules through the lane management unit to generate the trajectory. Figure 2When determining the policy trajectories of vehicles 201 and 203 in the scene, the policy selection is first performed on the respective vehicle control modules of vehicles 201 and 203, and then the vehicle control modules of vehicles 201 and 203 are called respectively. In one call, each vehicle control module calculates the vehicle state of the next time frame based on the information of the controlled vehicle using a physical model. The specific formula is as follows:

[0061]

[0062] Where Δt represents the time interval between adjacent time frames, p t The coordinates are two-dimensional, representing the vehicle's current position, v. t Let o be a two-dimensional vector representing the vehicle's current speed. t In radians, a represents the vehicle's current orientation. t The value is in radians, representing the current steering wheel angle of the vehicle; l represents the distance between the front and rear axles; p t+1 The coordinates are two-dimensional, representing the vehicle's position in the next time frame, o t+1 The value is in radians, representing the vehicle's orientation in the next time frame. This process is iterated N times to generate a strategy trajectory for N consecutive time frames. The trajectory generation unit then uses the trajectory information p... t+1 ,p t+2 …,p t+N Stored in the computer's storage medium.

[0063] Furthermore, the high-precision map module includes a map information storage unit and a map information query unit;

[0064] The map information storage unit includes a storage medium in a computer for storing map information, including the coordinate information of lane center lines in the map. Any lane center line is composed of a two-dimensional point coordinate sequence, which is represented as a polyline formed by connecting the points in the sequence in sequence.

[0065] The map information query unit is implemented using a gridded data structure to support efficient query operations on map information, including querying the coordinates of the center line of a specified lane, querying the adjacent lane information of the lane, and querying the center line of the lane closest to the coordinates of a given vehicle.

[0066] In such Figure 3 In the simulation scenario shown, one obstacle-avoiding vehicle is stationary, and the remaining vehicles are controlled by a system based on PID control that generates multi-strategy trajectories from a high-precision map. The simulation is recorded on video. The results demonstrate that a normal lane-changing and obstacle-avoidance trajectory can be generated.

Claims

1. A system for generating multi-strategy trajectories in high-precision maps based on PID control, characterized by, According to the physical model and heuristic rules, a plurality of strategy trajectories are generated, which are close to the real scene; Specifically, it includes a vehicle control module, a lane control module, a global control module and a high-precision map module; The global control module is connected with a plurality of lane control modules, the lane control module is connected with a plurality of vehicle control modules, and the high-precision map module is connected with the global control module; The above-mentioned modules are deployed in the computer with the system, and are called by the simulation test software; Wherein: The vehicle control module is used for controlling the trajectory generation of a vehicle, including maintaining the basic state of the vehicle and the information of the lane where the vehicle is located, the vehicle state information including coordinates, speed, orientation angle, steering wheel angle, target speed, and the lane where the vehicle is located, strategy information, obtaining the vehicle information before and after the vehicle from the lane control module and the vehicle information of the adjacent lane, and generating the future state and trajectory of the vehicle based on the obtained information and PID control method; The lane control module is used for controlling the vehicle control module corresponding to the vehicle on a lane, maintaining the front and rear sequence information of the vehicle on the lane, providing information interaction between different vehicles for the controlled vehicle control module, and supporting the interaction between the controlled vehicle control module and the global control module; The global control module is used for controlling all lane control modules, obtaining the coordinate information of the lane center line from the high-precision map module, maintaining the basic state and strategy information of all vehicles, and recording the generated trajectory of all vehicles; The high-precision map module is used for importing high-precision map information, using a data structure to store the information of the lane center line in the map information, supporting the positioning query of the vehicle coordinates in the lane, providing the point sequence representation of the center line of the lane and the adjacent lane information of the lane; The vehicle control module specifically includes: A vehicle speed PID control unit is configured to obtain a current speed of the vehicle at a current time frame t v t , and a target speed at the current time frame t , calculate a current speed , calculate a speed compensation value using a PID algorithm according to a difference between the current speed and the target speed , and calculate a speed at a next time frame s t+1 , wherein coefficients of proportional, integral, and differential terms of the PID algorithm are adjustable control parameters, denoted as S p , S i , and S d , respectively.​​ a vehicle steering wheel angle PID control unit for obtaining the current speed t v t , position coordinates p t , heading o t , steering wheel angle a t , a followed lane center line c t , a future position coordinates of the vehicle after 5 time frames based on the current state is calculated according to the following model:​ ， （1） wherein, Delta t is a time interval between adjacent time frames, is a predicted future position coordinate after 5 time frames, o t is an orientation angle of the vehicle, v t is a current speed of the vehicle, p t is a current position of the vehicle; Then, the current position of the vehicle is found using the following equation p t the nearest point position to the center line of the lane being followed c t the nearest point position to the center line of the lane being followed p c the nearest point position to the center line of the lane being followed ， （2） Among them, the lane center line is composed of m The sequence of two-dimensional point vectors is denoted as . ; d Distance from the center line of the lane p t The endpoint number of the nearest line segment; DIS ( X, Y, Z ) represents a point X From Y, Z The distance between two points on a line segment; CP ( X, Y, Z ) represents a point X From Y, Z Find the nearest point on the line segment formed by connecting the two points; then, use the following formula to calculate the point from... p c Starting from the current location, travel the corresponding distance along the center line of the lane to reach the future target coordinates. : ， （3） wherein the lane centerline is composed of a sequence of two-dimensional point vectors, denoted as m , , Delta t is the time interval between adjacent time frames, v t is the current speed of the vehicle, l is the length of the driving path of the vehicle in the next 5 time frames; d is the lane centerline c t the distance between the current position of the vehicle and the end point of the nearest line segment, p t the index of the end point of the nearest line segment, L i denotes the length of the driving path from p t to the point ; e denotes L d , L d+1 … L m-1 the maximum index satisfying the length of no more than l ; denotes the future target coordinate reached by advancing from p c to the point l along the lane centerline with a length of ; thereafter, the ratio of the distance between the future position coordinate and the future target coordinate to the current speed is taken as the gap, and the compensation value of the steering wheel angle is calculated by using the PID algorithm, so as to calculate the steering wheel angle of the next time frame a t+1 , wherein the coefficients of the proportional, integral and differential terms of the PID algorithm are adjustable control parameters, denoted as A p , A i , A d ; The strategy trajectory generation unit is used for obtaining the position and speed information of the vehicle around the current vehicle from the lane control module, selecting the acceleration, deceleration, straight driving, left and right lane changing strategy for the current vehicle, adjusting the target speed and target lane center line of the vehicle speed PID control unit and vehicle steering wheel angle PID control unit, and generating the corresponding strategy trajectory; Specifically: For the acceleration / deceleration strategy, the current target speed of the vehicle speed PID control unit is adjusted to implement, the target speed is replaced with where Delta s is the target speed increment, which indicates an acceleration strategy when Delta s is positive, and a deceleration strategy when Delta s is negative; For the keep straight strategy, no adjustment is made; for the left and right lane change strategy, the current followed lane center line of the steering wheel angle PID control unit of the vehicle is adjusted c t to achieve, with c t the left or right lane center line to replace the followed lane center line; For selecting the current strategy of the vehicle, the heuristic rules are used to judge according to the driving state of the vehicle and the information of the surrounding vehicles; When the speed of the front vehicle of the vehicle is less than the current speed of the vehicle and there is a risk of rear-end collision, the deceleration or lane changing strategy is selected; When the speed of the front vehicle of the vehicle is greater than the current speed of the vehicle, the acceleration or straight driving strategy is selected; If there are a plurality of feasible strategies under the current state, randomness is introduced to select the strategy.

2. The system for generating multi-strategy trajectories in a high-precision map based on PID control according to claim 1, wherein, The lane control module includes: The vehicle management unit is used for managing the vehicle control module corresponding to the vehicle on the lane, maintaining the front and rear sequence information of the vehicle, and supporting the operations of adding, deleting and querying the vehicle information on the lane; A lane-to-lane information interaction unit is configured to support information interaction between different lane control modules, maintain a linked list structure for each lane to maintain the numbers of left and right adjacent lanes, thereby supporting vehicle information interaction between adjacent lanes, and assist the lane control module in generating a lane-changing strategy trajectory.

3. The system for generating multi-strategy trajectories in a high-precision map based on PID control according to claim 2, characterized in that, The global control module comprises: A lane management unit is configured to manage all lane control modules, support creation, deletion, and query operations of the lane control modules, and perform information interaction between the lane control modules and all vehicle control modules; A trajectory generation unit is configured to call all vehicle control modules through the lane management unit, select a strategy for all vehicle control modules when generating a trajectory, and then call each vehicle control module in sequence; in one call, each vehicle control module calculates a vehicle state in a next time frame according to vehicle information controlled by the vehicle control module by using a physical model, and the specific formula is as follows: ， （4） wherein, Δt denotes a time interval between adjacent time frames, p t denotes a current position of the vehicle, v t denotes a current speed of the vehicle, o t denotes a current orientation angle of the vehicle, a t denotes a current steering wheel angle of the vehicle, l denotes a distance between front and rear axles of the vehicle, p t+1 denotes a position of the vehicle for a next time frame, o t+1 denotes an orientation of the vehicle for the next time frame; thereafter, the above process is iterated N times, i.e., a policy trajectory for consecutive N time frames is generated, the trajectory generation unit stores the trajectory information p t+1 , p t+2 … , p t+N in a storage medium of the computer.

4. The system for generating multi-strategy trajectories in a high-precision map based on PID control according to claim 3, characterized in that, The high-definition map module comprises: A map information storage unit comprises a storage medium in a computer and is configured to store map information, including coordinate information of a lane center line in a map, and any lane center line is composed of a two-dimensional point coordinate sequence and is represented as a polyline formed by sequentially connecting points in the sequence; A map information query unit is implemented by a gridding data structure and is configured to support efficient query operations on map information, including query of coordinate information of a lane center line of a specified lane, query of adjacent lane information of a lane, and query of a lane center line closest to a given vehicle coordinate.

5. The system for generating multi-strategy trajectories in high-precision maps based on PID control according to claim 4, characterized in that, The specific working process is as follows: (1) Obtain initialization information of a vehicle as system input, including coordinates, speed, orientation angle, steering wheel angle, target speed, and strategy information of the vehicle; (2) Import map information, including coordinate information of a lane center line, by using the high-definition map module, and locate each vehicle in a lane; (3) Create lane control modules for all lanes by using the global control module, create vehicle control modules for each vehicle by using the lane control modules, and manage the vehicle control modules to generate a strategy trajectory for each vehicle in a next time frame; (4) Generate a strategy trajectory for each vehicle for a continuous period of time by using the global control module for multiple iterations.

Images