Automatic driving vehicle path planning method

Abstract

The invention discloses an automatic driving vehicle path planning method, which is based on a lane map, the lane map comprises map point data and lane line data, and the formats of the map point data and the lane line data comprise road topology, passing point information, lane topology bridging lines, station parking time configuration and other multivariate information. And generating a lane map and a full-coverage track map. The vector map covering the whole road surface is generated through manual pre-recording data, the passing points are published, and according to the sequence of the passing points, the method can be applied to automatic driving logistics vehicles, is compatible with the complex requirements of scheduling operation of various vehicles such as unmanned sweeping vehicles, security vehicles and logistics vehicles, realizes that the automatic driving vehicles walk according to map tracks, and improves the scheduling efficiency. Station scheduling of the logistics vehicle is achieved, multi-circle sweeping work of the sweeping vehicle is achieved, and path planning is achieved to cover the whole road surface. And the method has great significance in playing a great role in deployment and application of an unmanned driving system in an actual customer scene.

Description

Technical Field

[0001] The present invention relates to the field of autonomous driving technology, and more specifically to a method for path planning of an autonomous driving vehicle. Background Art

[0002] With the development of autonomous driving technology, unmanned vehicles and equipment systems are being promoted and used more and more, including unmanned sweepers, unmanned security vehicles, unmanned logistics vehicles, unmanned shuttle vehicles, etc., which can create and improve various new quality productivity, greatly reduce the labor intensity of people in all walks of life such as urban cleaning, security, logistics, etc., improve the level of urban intelligence, and create a better and more convenient new life.

[0003] Path planning is the key to realize vehicle autonomous driving. At present, the existing path planning of autonomous driving vehicles on structured roads, such as the Chinese patent with publication number CN113219985A discloses a path planning method, device, and sweeper for a sweeper; the Chinese patent with publication number CN118348974A discloses a path planning method and system for an intelligent sweeper based on multi-level planning; and the Chinese patent with publication number CN116380095A discloses a path planning method, device, equipment, vehicle, and medium for autonomous driving vehicles. These patents mainly focus on solving problems such as obstacle avoidance optimization, safety, and smoothness of path planning. Summary of the invention

[0004] The present invention provides a path planning method for autonomous driving vehicles, which realizes path planning and site scheduling for autonomous driving logistics vehicles and security vehicles, and multi-circle planning for autonomous driving road sweepers to achieve full coverage of the road surface.

[0005] The present invention adopts the following technical solution: A method for path planning of an autonomous driving vehicle is based on a lane map, wherein the lane map includes map point data and lane line data, and comprises the following steps: S1. Pre-record map point data and lane line data to generate a lane map and a full coverage trajectory map; the format fields of the map point data include longitude, latitude, flag 1, flag 2, Utm coordinate X and Utm coordinate Y; the format fields of the lane line data include lane ID, next lane ID, left lane ID, right lane ID, left center line ID, right center line ID, lane type, lane mode 1, lane mode 2, intersection ID, road direction, maximum speed, left boundary ID, right boundary ID and road width; S2. Create a waypoint route, where the data format fields of the waypoint route include a route ID, a route name, a waypoint ID list, and an adapted vehicle model; S3: Route planning based on waypoints, line station scheduling, and high-coverage driving planning.

[0006] In a preferred implementation scheme, the lane map generation method in the above step S1 is specifically as follows: S11, reading the longitude and latitude data of the vehicle position, saving the longitude and latitude data at a distance of 20 cm to form the center line of the lane; S12, using laser radar to create a point cloud map of the vehicle operating environment; S13, widening the lane with a set lane width to obtain the lane boundary line; S14, performing local trimming of the lane boundary line according to the point cloud map; S15, marking the relationship between the marked lanes, including the left and right lanes, and the front and rear lane IDs, and generating a map topological relationship; S16, saving the generated planning topology map.

[0007] In a preferred implementation scheme, the coding rule of the waypoint ID in the above step S2 is that each waypoint is connected with a semicolon and a "-" sign, the station ID before the semicolon indicates that the vehicle needs to stop at this location, and the "-" sign connects the waypoints.

[0008] In a preferred embodiment, the path planning method based on waypoints in the above step S3 includes the following steps: S311, read the waypoint data into the list pass_vec, read the lane map and put it into the map_list linked list according to the lane relationship; S312, read the current position of the vehicle from the high-precision positioning module; S313, calculate the lane L0 closest to the current position as the starting lane, and set the waypoint data list reading pointer i=0; S314, read the i-th data in the waypoint data list; S315, calculate the distance from the current viapoint i The nearest lane is used as the destination lane; S316, traverse the lane map data map_list linked list, obtain all the lanes from the starting lane to the destination lane and add them to the lane list pass_vec[i], and connect them to the total lane list passLane_all; S317, determine whether it is the last waypoint. If it is not the last waypoint, the waypoint reading count is increased by 1 and returns to S314, and the starting lane is set as the current destination lane; S318, if it is the last waypoint, it means that the traversal search is completed, then passLane_all is the final planned path obtained, and the total number of all lane points on the passLane_all lane list represents the distance of the entire planned path.

[0009] In a preferred implementation scheme, the route station scheduling method in the above step S3 specifically includes: S321, reading the waypoint route file, and reading the parking time configuration of each station according to the current execution task of the autonomous driving vehicle; S322, receiving the route file and dividing the point set into multiple waypoint subsets with a semicolon; S323, extracting the i-th subset of the waypoint subset for planning according to the waypoint-based planning method; S324, controlling the vehicle to drive after issuing the planned path; S325, after the vehicle arrives at the parking station, it stops and waits according to the parking data of the station; S326, the vehicle arrives at the path switching station before the semicolon separator to stop, and after the on-site scheduling task is completed, the next waypoint point set is extracted for path planning until the end point is reached and the task is completed.

[0010] In a preferred implementation scheme, the high coverage driving planning method in the above step S3 specifically includes: S331, firstly collecting the full coverage trajectory lines of the inner, middle and outer circles to obtain the inner circle, middle circle and outer circle three trajectory lines; S332, designing bridge connection lines for the three trajectory lines; S333, setting waypoints P1, P2, P3, P4, P5 and P6 on the three trajectory lines, starting from any point on the trajectory line, if these waypoints are traversed in sequence, all routes can be covered, that is, a traversal waypoint set is obtained: Loop_pass_set = [P1, P2, P3, P4, P5, P6]; S334, get the current position of the vehicle, set it as Pnow, use the waypoint-based planning method to plan the path from the current position Pnow to all the waypoints (P1, P2, P3, P4, P5, P6) in sequence, and the number of points on the planned path can be used to determine which waypoint the current position is closest to; S335, combine the current closest waypoint P2 to generate the current driving waypoint point set pass_set according to the preset waypoint set Loop_pass_set, and use the preset point set as a circular queue to generate the current waypoint point set; S336, use the waypoint-based planning method to plan in sequence and use the current point set pass_set for path planning to obtain a pass_round path covering all paths; S337, according to the point closest to the current position on the adjacent circle as the dividing point, split the pass_round path into three circles of Round1, Round2, and Round3, and drive the automatic driving in sequence.

[0011] It can be seen from the above description of the present invention that, compared with the prior art, the present invention has the following advantages: The map point data and lane line data formats of the present invention include multiple information such as road topology, waypoint information, lane topology bridge lines, and site parking time configuration, which can be used as a basis for path planning. A vector map covering the entire road surface is generated by manually pre-recording data, and waypoints are published. According to the order of waypoints, the autonomous driving vehicle can move along the map trajectory, the site scheduling of the logistics vehicle can be realized, and the multiple-circle cleaning work of the sweeper can be realized, so that the path planning covers the entire road surface. The present invention does not require additional sensors to identify the road surface, and the hardware and software costs are relatively low. The environmental applicability is also high, and it is easy to simulate the planning effect of a real vehicle. It can be applied to autonomous driving logistics vehicles, and is compatible with the complex needs of scheduling operations of various vehicles such as unmanned sweepers, security vehicles, and logistics vehicles. It is of great significance to the deployment and application of unmanned driving systems in actual customer scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] Figure 1 A schematic diagram of a lane map generated by an embodiment of the present invention.

[0013] Figure 2 Schematic diagram of lane topology according to an embodiment of the present invention.

[0014] Figure 3 Schematic diagram of a full coverage trajectory map according to an embodiment of the present invention.

[0015] Figure 4 The figure is a flow chart of path planning for waypoints according to an embodiment of the present invention.

[0016] Figure 5 A map route schematic diagram of route station scheduling according to an embodiment of the present invention.

[0017] Figure 6 This is a flow chart of route station scheduling according to an embodiment of the present invention.

[0018] Figure 7 Schematic diagram of three trajectory lines planned for high-coverage driving according to an embodiment of the present invention.

[0019] Figure 8 This is a flow chart of high-coverage driving planning according to an embodiment of the present invention. DETAILED DESCRIPTION

[0020] The specific implementation of the present invention is described below. In order to fully understand the present invention, many details are described below, but for those skilled in the art, the present invention can be implemented without these details. For well-known components, methods and processes, they are not described in detail below.

[0021] A method for path planning of an autonomous driving vehicle is based on a lane map, wherein the lane map includes map point data and lane line data, and comprises the following steps: 1. Pre-record map point data and lane line data to generate lane maps and full coverage trajectory maps.

[0022] The map point data format is as follows: Serial number Fields illustrate 1 longitude 2 latitude 3 Flag 1 Indicates the credibility of the current longitude and latitude 4 Flag 2 Reserved fields 5 Utm coordinate X Project longitude and latitude into UTM coordinates 6 Utm coordinate Y Project longitude and latitude into UTM coordinates The text files are stored as follows: 117.9672122200 24.5931633600 2 3 597923.09393755 2720244.52946340 117.9672072062 24.5931585475 2 3 597922.59003980 2720243.99299438 117.9672030734 24.5931545479 2 3 597922.17470988 2720243.54716692 117.9671993273 24.5931507783 2 3 597921.79835397 2720243.12708313 117.9671955162 24.5931468850 2 3 597921.41551309 2720242.69325545 117.9671917600 24.5931430862 2 3 597921.03815724 2720242.26993110 117.9671882925 24.5931395811 2 3 597920.68980344 2720241.87933440 117.9671847225 24.5931359232 2 3 597920.33119048 2720241.47174479 117.9671811597 24.5931322322 2 3 597919.97333224 2720241.06049506 117.9671770523 24.5931279015 2 3 597919.56083147 2720240.57802208 117.9671729800 24.5931235000 2 3 597919.15193959 2720240.08773418 117.9671693567 24.5931195678 2 3 597918.78814337 2720239.64973266 117.9671659243 24.5931157081 2 3 597918.44361915 2720239.21989508 117.9671624000 24.5931116000 2 3 597918.08998338 2720238.76248608 117.9671591885 24.5931076631 2 3 597917.76788508 2720238.32425708 Each type of data is separated by "\t".

[0023] Lane line data format: Serial number Fields illustrate 1 Lane ID 2 Next lane ID 3 Left Lane ID 4 Right Lane ID 5 Left Centerline ID 6 Right Centerline ID 7 Road Type Intersections, zebra crossings, stop lines, etc. 8 Lane Type Straight roads, curves, etc. 9 Lane Mode 1 10 Lane Mode 2 11 Intersection ID 12 Road Direction Turn left, turn right 13 Maximum speed 14 Left border ID 15 Right border ID 16 Road width The text files are stored as follows: 48 3 0 0 2 0 3 0 4 3 0 0 10 18 19 0.000 3 4 0 0 0 0 3 0 2 3 0 0 10 20 28 0.000 4 5,16 0 0 0 0 3 0 4 3 0 0 10 45 46 0.000 5 11 0 0 16 0 3 0 4 3 0 0 12 43 44 0.000 11 7 0 21 14 0 3 0 2 3 0 0 20 39 40 0.000 7 23 0 0 13 0 3 0 4 3 0 0 12 59 60 0.000 23 9,1 0 0 61 0 3 0 2 3 0 0 12 30 31 0.000 9 24 0 0 6 0 3 0 4 3 0 0 10 27 25 0.000 24 48 0 0 1 0 3 0 2 3 0 0 15 33 34 0.000 22 2 0 0 0 12 3 0 2 3 0 0 15 29 33 0.000 6 22 0 0 0 9 3 0 4 3 0 0 10 26 27 0.000 13 61 0 0 0 7 3 0 4 3 0 0 12 57 58 0.000 14 13 0 0 0 11 3 0 2 3 0 0 20 37 39 0.000 16 14 0 0 0 5 3 0 4 0 0 0 12 41 42 0.000 37 0 0 0 0 0 13 1 2 3 0 0 10 0 0 0.000 39 0 0 0 0 0 11 1 2 3 0 0 10 0 0 0.000 40 0 0 0 0 0 13 1 2 3 0 0 10 0 21 0.000 45 0 0 0 0 0 13 1 4 3 0 0 10 0 0 0.000 46 0 0 0 0 0 13 1 4 3 0 0 10 0 0 0.000 Each type of data is separated by "\t".

[0024] The road map generation method is as follows: (1) Read the longitude and latitude data of the vehicle position and save them at a distance of 20 cm to form the center line of the lane.

[0025] (2) In order to obtain more accurate lane line data, laser radar is used to create a point cloud map of the vehicle operating environment.

[0026] (3) Widen the lane to the set lane width to obtain the lane boundary line.

[0027] (4) Perform local trimming of lane boundaries based on the point cloud map to avoid special situations such as being too close to dangerous road boundaries.

[0028] (5) Mark the relationship between lanes, including left and right lanes, front and rear lane IDs, and generate map topology relationships.

[0029] (6) Save and generate the planning topology map.

[0030] Make a lane map generated like Figure 1 As shown in Figure 2, the lane topology is as follows: Figure 2 As shown, Figure 2 The yellow line in the middle is the center line of the lane, and the white lines generated by the equal distance offset inside and outside each yellow line are the lanes after topology.

[0031] For unmanned road sweepers, a full coverage track map can be created to achieve full coverage of the road by the vehicle, thus achieving a comprehensive cleaning effect. Figure 3 The middle yellow line is the center line of the lane, and the purple line is the full coverage trajectory route. Each lane center line contains 3 purple full coverage trajectory routes, which can enable the self-driving sweeper to achieve the maximum cleaning area during driving.

[0032] 2. Create a waypoint route The map is unique. Different routes can be set by configuring the waypoints. The route file data format is as follows: Serial number Fields illustrate 1 Line ID 2 Line Name 3 Waypoint ID list 4 Applicable models Adjust path planning decisions based on vehicle type The line file is as follows: Path ID Path name Waypoint information (ID is read from the stop file) Vehicle Type 40001 Test Center 28 _30 1 40002 Small circle positive 35_28 1 40003 Small circle reverse 31_24 1 40004 Big Circle 39_42 1 40005 Back to Test Center 59_60 1 20001 Trial Center Sites 16_19 1 20002 Small circle reverse site 20_23 1 20003 Small circle positive site 25；27 3 20004 Small Circle Sightseeing Bus Stop 24；26 2 The coding rule of the waypoint ID is to connect each waypoint with a semicolon and a "-". The station ID before the semicolon indicates that the vehicle needs to stop at this location. For example, "25;27" means that the vehicle automatically drives to the station numbered 25, stops, and can accept manual operations at this station, so it is a path planning switching point.

[0033] The “-” sign connects the waypoints, for example, “35-38” means that you need to pass through the two waypoints with ID numbers 35 and 38.

[0034] The vehicle type parameter can specify the type of autonomous driving vehicle, such as a logistics vehicle, a street sweeper, or a security vehicle.

[0035] 3. Path planning methods, including path planning methods based on waypoints, line station scheduling methods, and high-coverage driving planning methods.

[0036] 1. Path planning method based on waypoints, refer to Figure 4 , including the following steps: (1) Read the waypoint data into the list pass_vec, read the lane map and put it into the map_list list according to the lane relationship.

[0037] (2) Read the vehicle’s current location from the high-precision positioning module.

[0038] (3) Calculate the lane L0 closest to the current position as the starting lane and set the waypoint data list reading pointer i = 0.

[0039] (4) Read the i-th data in the waypoint data list.

[0040] (5) Calculate the lane closest to the current passing point i as the destination lane.

[0041] (6) Traverse the lane map data map_list linked list, obtain all the lanes from the starting lane to the destination lane and add them to the lane list pass_vec[i], and connect them to the total lane list passLane_all.

[0042] (7) Determine whether it is the last waypoint. If it is not the last waypoint, add 1 to the waypoint reading count and return to (4), and set the starting lane to the current destination lane.

[0043] (8) If it is the last waypoint, it means the traversal search is over, and passLane_all is the final planned path. The total number of all lane points on the passLane_all lane list can represent the distance of the entire planned path.

[0044] 2. Line station scheduling method For the application scenarios of autonomous driving vehicles, there is a common demand for site scheduling. The main purpose is to dispatch the vehicle to a specific location on the map and stop to handle special business as needed. For example, a logistics vehicle arrives at a certain location to load and unload goods, a security vehicle arrives at a specific location to take photos and collect evidence, and a street sweeper is dispatched to a designated location for automatic charging, garbage disposal and water addition, etc.

[0045] Reference Figure 5 , take the map route diagram in the figure as an example: The starting position of the autonomous driving car is P1, and there are two destinations, one is P12 destination, and the other is P9 destination; P4 is the point where the car must stop. After stopping at P4, the vehicle will be manually controlled (such as unloading the car). The parking time at other passing points is configured by the configuration file. To achieve this planning, this patent design has two key configuration files: The first is the route waypoint configuration file. According to the waypoint route creation method mentioned above, the main contents are as follows: route to end point 12: 1-2-3-4; 5-12; route to end point 9: 1-2-3-4; 5-6-8-9.

[0046] The second is the parking time profile. According to the needs of a certain scheduling task, the parking waiting time for certain parking spots can be set. For example, for Task 1, set: 2-80, 3-60, 6-90, which means that the automatic driving parking time for the three stations 2, 3, and 6 is 80 seconds, 60 seconds, and 90 seconds respectively, and the other stations do not need to stop. For Task 2, you can set: 2-60, 5-70, 7-90, which means that the automatic driving parking time for the three stations 2, 5, and 7 is 60 seconds, 70 seconds, and 90 seconds respectively, and the other stations do not need to stop.

[0047] Referring to the figure, the line station scheduling method specifically includes: (1) Read the waypoint route file and read the parking time configuration of each station according to the current execution task of the autonomous driving vehicle; (2) Receive the route file and use semicolons to split the point set into multiple waypoint subsets; (3) extracting the i-th subset of the waypoint subset for planning according to the waypoint-based planning method; (4) Control vehicle movement after issuing the planned path; (5) After the vehicle arrives at the parking station, it stops and waits according to the parking data of the station; (6) The vehicle stops at the path switching station before the semicolon separator and extracts the next set of passing points for path planning after the on-site scheduling task is completed, until it reaches the destination and completes the task.

[0048] 3. Use the point planning method to design a high coverage planning method to achieve full coverage cleaning design of the unmanned sweeper.

[0049] Reference Figure 7 , inner circle (Round1), middle circle (Round2), outer circle (Round3), three circles of autonomous driving can just cover the entire road. In order to ensure that autonomous driving can correctly cover three circles starting from any position, this embodiment adopts the following method: (1) First, collect the full coverage trajectory lines of the inner, middle and outer circles to obtain the three trajectory lines of the inner circle, middle circle and outer circle; (2) Design bridge connection lines for the three track lines; as shown in the figure below: L1 is the bridge line from the inner circle (Round1) to the middle circle (Round2), L2 is the bridge line from the middle circle (Round2) to the inner circle (Round1), L3 is the bridge line from the middle circle (Round2) to the outer circle (Round3), and L4 is the bridge line from the outer circle (Round3) to the middle circle (Round2). In order to improve the efficiency of path switching and avoid repeated routes, the bridge lines should be set as close as possible while ensuring that vehicles can pass.

[0050] (3) Set waypoints P1, P2, P3, P4, P5, and P6 on the three trajectory lines. Starting from any point on the trajectory line, if you traverse these waypoints in sequence, you can cover all routes, that is, you get a traversal waypoint set: Loop_pass_set = [P1, P2, P3, P4, P5, P6].

[0051] Set up the bridge line and waypoints, refer to Figure 8 , adopt the following planning method to achieve full coverage driving: (1) Get the current vehicle position and set it to Pnow.

[0052] (2) Use the waypoint-based planning method to plan the path from the current position Pnow to all the waypoints (P1, P2, P3, P4, P5, P6) in sequence. According to the number of points on the planned path, it can be determined which waypoint the current position is closest to, such as Figure 7The point closest to Pnow is P2.

[0053] (3) In combination with the current nearest waypoint P2, the current driving waypoint set pass_set is generated according to the preset waypoint set Loop_pass_set, and the preset point set is regarded as a circular queue to generate the current waypoint set.

[0054] (4) Use the waypoint-based planning method to plan in sequence and use the current point set pass_set for path planning to obtain a pass_round path that covers all paths.

[0055] (5) Using the point closest to the current position on the adjacent circle as the dividing point, the pass_round path is split into three paths: Round 1, Round 2, and Round 3, and the autonomous driving is performed in sequence.

[0056] The above is only a specific implementation of the present invention, but the design concept of the present invention is not limited thereto. Any non-substantial changes to the present invention using this concept shall be deemed as an infringement of the protection scope of the present invention.

Claims

1. A method for path planning of an autonomous driving vehicle, based on a lane map, wherein the lane map includes map point data and lane line data, characterized in that: The following steps are involved: S1. Pre-record map point data and lane line data to generate a lane map and a full coverage trajectory map; the format fields of the map point data include longitude, latitude, flag 1, flag 2, Utm coordinate X and Utm coordinate Y; the format fields of the lane line data include lane ID, next lane ID, left lane ID, right lane ID, left center line ID, right center line ID, lane type, lane mode 1, lane mode 2, intersection ID, road direction, maximum speed, left boundary ID, right boundary ID and road width; S2. Create a waypoint route, where the data format fields of the waypoint route include a route ID, a route name, a waypoint ID list, and an adapted vehicle model; S3: Route planning based on waypoints, line station scheduling, and high-coverage driving planning.

2. A method for path planning of an autonomous driving vehicle according to claim 1, characterized in that: The lane map generation method in step S1 is specifically as follows: S11, reading the longitude and latitude data of the vehicle position, saving the longitude and latitude data at a distance of 20 cm to form the lane centerline; S12, using laser radar to produce a point cloud map of the vehicle operating environment; S13, widening the lane with a set lane width to obtain the lane boundary line; S14, performing local trimming of the lane boundary line according to the point cloud map; S15, marking the relationship between the marked lanes, including the left and right lanes, and the front and rear lane IDs, and generating a map topological relationship; S16, saving and generating a planning topological map.

3. The method for path planning of an autonomous driving vehicle according to claim 1, characterized in that: The coding rule of the waypoint ID in step S2 is that each waypoint is connected by a semicolon and a "-" sign, the station ID before the semicolon indicates that the vehicle needs to stop at this location, and the "-" sign connects the waypoints.

4. The method for path planning of an autonomous driving vehicle according to claim 1, wherein: The path planning method based on waypoints in step S3 includes the following steps: S311, reading waypoint data into list pass_vec, reading lane map and putting it into map_list linked list according to lane relationship; S312, reading the current position of the vehicle from the high-precision positioning module; S313, calculating the lane L0 closest to the current position as the starting lane, setting the waypoint data list reading pointer i=0; S314, reading the i-th data in the waypoint data list; S315, calculating the distance from the current viapoint i The nearest lane is used as the destination lane; S316, traverse the lane map data map_list linked list, obtain all the lanes from the starting lane to the destination lane and add them to the lane list pass_vec[i], and connect them to the total lane list passLane_all; S317, determine whether it is the last waypoint. If it is not the last waypoint, the waypoint reading count is increased by 1 and returns to S314, and the starting lane is set as the current destination lane; S318, if it is the last waypoint, it means that the traversal search is completed, then passLane_all is the final planned path obtained, and the total number of all lane points on the passLane_all lane list represents the distance of the entire planned path.

5. The method for path planning of an autonomous driving vehicle according to claim 1, characterized in that: The route site scheduling method in step S3 specifically includes: S321, reading the waypoint route file, and reading the parking time configuration of each site according to the current execution task of the autonomous driving vehicle; S322, receiving the route file and dividing the point set into multiple waypoint subsets with a semicolon; S323, extracting the i-th subset of the waypoint subset for planning according to the waypoint-based planning method; S324, controlling the vehicle to drive after issuing the planned path; S325, after the vehicle arrives at the parking site, it stops and waits according to the site parking data; S326, the vehicle arrives at the path switching site before the semicolon separator to stop, and extracts the next waypoint point set for path planning after the on-site scheduling task is completed, until the end point is reached and the task is completed.

6. The method for path planning of an autonomous driving vehicle according to claim 1, characterized in that: The high coverage driving planning method in step S3 specifically includes: S331, first collecting full coverage trajectory lines of the inner, middle and outer circles to obtain three trajectory lines of the inner circle, the middle circle and the outer circle; S332, designing bridge connection lines for the three trajectory lines; S333, setting waypoints P1, P2, P3, P4, P5 and P6 on the three trajectory lines, starting from any point on the trajectory line, if these waypoints are traversed in sequence, all routes can be covered, that is, a traversal waypoint set is obtained: Loop_pass_set = [P1, P2, P3, P4, P5, P6]; S334, get the current position of the vehicle, set it as Pnow, use the waypoint-based planning method to plan the path from the current position Pnow to all the waypoints (P1, P2, P3, P4, P5, P6) in sequence, and the number of points on the planned path can be used to determine which waypoint the current position is closest to; S335, combine the current closest waypoint P2 to generate the current driving waypoint point set pass_set according to the preset waypoint set Loop_pass_set, and use the preset point set as a circular queue to generate the current waypoint point set; S336, use the waypoint-based planning method to plan in sequence and use the current point set pass_set for path planning to obtain a pass_round path covering all paths; S337, according to the point closest to the current position on the adjacent circle as the dividing point, split the pass_round path into three circles of Round1, Round2, and Round3, and drive the automatic driving in sequence.

Images