A real-time path planning method and AGV

Abstract

The present invention discloses a real-time path planning method, comprising the following steps: S1, constructing a topological map of a vehicle driving environment; S2, detecting whether the current position P0 of the vehicle is on a fixed path, if the detection result is no, executing step S3; S3, determining the station P1 closest to the current position on the fixed path, and planning the shortest path R0 from the station P1 to the task station; S4, calculating the angle θ between the current position P0 and the station P1, if abs(θ)>90°, executing step S5, S5, taking the next station P2 of the station P1 in the path R0 as the starting point, planning the shortest path R1 from the station P2 to the task node Ps, and then planning the shortest driving path R2 from the current position P0 to the station P2, the splicing of the path R2 and the path R1 forms the driving path R from the current position P0 to the task station Ps. When the vehicle deviates from the fixed driving path, searching for a station on the driving path that is close to the current position and has an angle less than 90 degrees for shortest path planning, so as to avoid the vehicle from turning around.

Description

Technical Field

[0001] The present invention belongs to the field of robot technology, and more specifically, the present invention relates to a real-time path planning method and an AGV. Background Art

[0002] With the development of material conveying systems, flexible manufacturing systems, and automated warehouse systems, AGVs (Automated Guided Vehicles) and vehicles, as important components of material conveying systems and flexible manufacturing systems, can solve the drawbacks of traditional logistics systems and have been widely used in the transportation links of the manufacturing industry. In actual industrial sites, there are often multiple AGVs operating at the same time. When these AGVs share a fixed path, an effective navigation strategy determines whether each robot can effectively coordinate operations. The time complexity and robustness of the path planning algorithm determine whether the robot can search for an effective path without colliding with other robots and obstacles.

[0003] Nowadays, path planning methods such as the A* algorithm use the A* algorithm to construct a heuristic function and set a reasonable evaluation function to solve the driving path of the AGV task. However, this method is only suitable for path planning when the AGV is in a node path. A certain section of the AGV driving path may suddenly be occupied by other obstacles. In addition, the AGV needs to avoid obstacles. The AGV deviates from the node path and cannot travel on the path planned by the A* algorithm. Summary of the invention

[0004] The present invention provides a real-time path planning method, aiming to improve the above-mentioned problem.

[0005] The present invention is implemented as follows: a real-time path planning method, the method specifically comprising the following steps:

[0006] S1. Build a topological map of the vehicle driving environment and generate the shortest driving path from the starting site to the task site;

[0007] S2, during the vehicle driving process, real-time detection of whether the current position P0 of the vehicle is on a fixed path, if the detection result is no, then executing step S3;

[0008] S3, determine the station P1 closest to the current location on the fixed path, and plan the shortest path R0 from station P1 to the task station;

[0009] S4, calculate the angle θ between the current position P0 and the site P1, if abs(θ)>90°, execute step S5,

[0010] S5. Take the next site P2 of site P1 in path R0 as the starting point, plan the shortest path R1 from site P2 to task node Ps, and then plan the shortest driving path R2 from the current position P0 to site P2. The connection of path R2 and path R1 forms the driving path R from the current position P0 to the task site Ps.

[0011] Further, if abs(θ)≤90°, execute step S6;

[0012] S6. Directly plan the shortest driving path R3 from site P1 to task site Ps, and then plan the shortest driving path R4 from current position P0 to site P1. The concatenation of path R3 and path R4 forms the driving path R from current position P0 to task site.

[0013] Furthermore, after step S6, the method further includes:

[0014] S7. After the path planning is completed, the road section weight of the current vehicle is displayed, and the length of each road section in the planned path is updated based on the road section weight of the vehicle, wherein the updated road section length is equal to the sum of the initial length of the road section and the road section weight*the initial length of the road section.

[0015] Furthermore, during the driving process of the vehicle based on the planned path, it is detected in real time whether the current position of the vehicle is the station location. If the current position is the station location, the number of vehicles on the next driving section is determined. If the number of vehicles on the next section is greater than the number threshold, the next section is marked as blocked, and the current position is used as the starting point to re-plan the shortest driving path from the current position to the task site. The vehicle drives based on the re-planned shortest driving path, and step S7 is executed for the re-planned shortest driving path.

[0016] Furthermore, the calculation formula of the angle θ between the current position P0 and the site P1 is as follows:

[0017]

[0018] Among them, angle is the direction of the vehicle's current position, (x0, y0) is the coordinate of the current position P0 in the grid map, and (x1, y1) is the coordinate of the site P1 in the grid map.

[0019] The present invention is implemented as follows: an AGV, wherein the AGV includes a processor and a memory, wherein the memory stores at least one instruction, at least one program, a code set or an instruction set, and the at least one instruction, the at least one program, the code set or the instruction set are loaded and executed by the processor to implement the above-mentioned real-time path planning method.

[0020] The present invention proposes a real-time path planning method with the following beneficial technical effects:

[0021] 1) When the vehicle deviates from the fixed driving path, it searches for a station on the driving path that is close to the current position and has an angle less than 90 degrees to plan the shortest path to avoid the vehicle turning around;

[0022] 2) Path length update: The more vehicles that pass through a path, the longer its length in the grid map will be. When performing global shortest path planning, the probability of it being selected as a driving section will become smaller, thereby reducing the probability of road congestion.

[0023] 3) Monitor the number of vehicles on the next road section in real time. When there are too many vehicles on the next road section, re-plan the shortest path based on the current position and drive based on the secondary planned driving path to reduce the traffic volume on the next road section and reduce the risk of vehicle congestion on the next road section;

[0024] To sum up, the path planning method proposed in the present invention has a wide range of applications and is suitable for path planning where multiple vehicles are operating simultaneously within a park factory. It optimizes the original scheduling system and solves the scenario where AGV or laser forklifts cannot turn around. It solves the problem of planning the shortest path at any vehicle position, reduces the difficulty of the system, and improves transportation efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 A flow chart of a real-time path planning method provided by an embodiment of the present invention;

[0026] Figure 2 A schematic diagram of a topological map of a driving environment provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0027] The specific implementation modes of the present invention are further explained in detail below by describing the embodiments with reference to the accompanying drawings, so as to help those skilled in the art to have a more complete, accurate and in-depth understanding of the inventive concept and technical solution of the present invention.

[0028] Figure 1 A flowchart of a real-time path planning method provided by an embodiment of the present invention, the method specifically comprises the following steps:

[0029] S1. Build a topological map of the vehicle driving environment and directly plan the shortest driving path from the starting station to the task station based on the AStar algorithm;

[0030] Design a route map for vehicles in the factory, mark the coordinates of the vehicle's charging stations, loading / unloading stations, standby stations, and route stations. The vehicle's route is a two-way route, and the vehicle can only accept one task at the same time period. While the vehicle is executing the current task, no task can be assigned to the vehicle. Use a topological structure to establish a working environment model for AGV or lidar in the material transportation workshop in the factory. The stations in the environment are the nodes in the topological map, and the edges between two nodes are the paths.

[0031] S2. During the driving process, the vehicle detects in real time whether the current position P0 of the vehicle is on a fixed path. If the detection result is no, step S3 is executed. If the detection result is yes, the shortest path from the current position P0 to the task node Ps is planned based on the AStar algorithm.

[0032] The vehicle in the present invention is an AGV or a laser forklift, and the vehicle can obtain its own position information in real time; and save the vehicle status information in a database in real time, the status information includes the current vehicle number, the vehicle position, the vehicle angle, the vehicle status, whether the vehicle is idle, and the vehicle target position, wherein the current status includes: stop, drive and turn;

[0033] The "fixed path" in the present invention refers to a fixed driving path between two sites, that is, an edge between two nodes in a topological map. The vehicle loads materials at the starting site. After loading, the vehicle drives along the driving path composed of the fixed path to the task site, where the material is unloaded. Figure 2 For illustration, the P1 station is the starting station, the P5 station is the task station, the fixed driving path is the road section between the P1 station and the P4 station, and the road section between the P4 station and the P5 station is a fixed path. When the vehicle avoids a sudden obstacle, the vehicle may deviate from the fixed path. At this time, the vehicle path needs to be planned based on the real-time path planning method provided by the present invention.

[0034] S3, determine the station P1 closest to the current location on the fixed path, and plan the shortest path R0 from station P1 to the task station based on the AStar algorithm;

[0035] S4, calculate the angle θ between the current position P0 and the site P1, if abs(θ)>90°, execute step S5, if abs(θ)≤90°, execute step S6;

[0036] In the embodiment of the present invention, the current posture of the vehicle is received, which is composed of the position P0 (x0, y0) and the direction (angle). Assuming that the coordinates of the current position P0 in the grid map are (x0, y0), and the coordinates of the station P1 in the grid map are (x1, y1), then the calculation formula of the angle θ between the current position P0 and the station P1 is as follows:

[0037]

[0038] Wherein, angle is the direction of the vehicle's current position. When the angle θ is greater than 90 degrees, it indicates that the station P1 is at the rear of the driving path. If the station P1 is included in the planned path, the vehicle will need to turn around. In order to avoid the vehicle turning around, the angle θ between the current position P0 and the station P1 needs to be less than 90 degrees.

[0039] S5, taking the next site P2 of site P1 in path R0 as the starting point, planning the shortest path R1 from site P2 to task node Ps based on AStar algorithm, and then planning the shortest driving path R2 from current position P0 to site P2 based on AStar algorithm, the concatenation of path R2 and path R1 forms the driving path R from current position P0 to task site Ps;

[0040] S6. Directly plan the shortest driving path R3 from site P1 to task site Ps based on the AStar algorithm, and then plan the shortest driving path R4 from the current position P0 to site P1 based on the AStar algorithm. The concatenation of path R3 and path R4 forms the driving path R from the current position P0 to the task site.

[0041] In the embodiment of the present invention, after step S6, the method further includes:

[0042] S7. After the path planning is completed, the section value of the current vehicle is updated, and the length of each section in the planned path is updated based on the section weight of the vehicle. For the sections in the planned path, the updated section length = initial section length + section weight * initial section length.

[0043] Assume that the planned path of vehicle A is composed of section 1, section 2 and section 3 in sequence, where the initial lengths of section 1, section 2 and section 3 are 5m, 8m and 10m respectively. Assume that the section weight of the current vehicle is 0.1, and the section weight is updated according to the set step size (for example, the step size is 0.1), and the section weight is updated to 0.2. Then the updated planned path is: section 1 is 6m long, section 2 is 9.6m long and section 3 is 12m long. The more vehicles pass through the path, the longer its length in the grid map will be. When performing global shortest path planning, the probability of it being selected as a driving section will become smaller, thereby reducing the probability of section congestion.

[0044] In an embodiment of the present invention, during the driving process, the vehicle detects in real time whether the current position of the vehicle is a station position. If the current position is a station position, the number of vehicles on the next driving section is determined. If the number of vehicles on the next section is greater than a number threshold, the next section is marked as blocked, and the current position is used as the starting point to re-plan the shortest driving path from the current position to the task station. The vehicle drives based on the re-planned shortest driving path, and step S7 is executed for the re-planned shortest driving path. If the number of vehicles on the next section is less than or equal to the number threshold, the vehicle continues to drive based on the initially planned path. If the current position is not a station position, the vehicle continues to drive based on the initially planned path.

[0045] The present invention detects the number of vehicles on the next road section in real time. When the number of vehicles on the next road section is too large, the next road section is marked as blocked, and the shortest path planning is re-performed based on the current position. There are two results of the secondary shortest path planning. One is that if there is no driving path from the current position to the task site, the vehicle drives based on the shortest driving path planned previously; the other is that if there is a driving path from the current position to the task site, the vehicle drives based on the secondary planned driving path to reduce the traffic volume of the next road section and reduce the risk of vehicle congestion in the next section.

[0046] The present invention reduces the risk of traffic jam to the greatest extent by updating the road section length during global planning and monitoring the number of vehicles on local paths.

[0047] An embodiment of the present application also provides an AGV, the AGV comprising: a processor and a memory, the memory storing at least one instruction, at least one program, a code set or an instruction set, the at least one instruction, the at least one program, the code set or the instruction set being loaded and executed by the processor to implement the above-mentioned real-time path planning method.

[0048] The present invention is described above by way of example in conjunction with the accompanying drawings. It is obvious that the specific implementation of the present invention is not limited to the above-mentioned method. As long as various non-substantial improvements are made using the method concept and technical solution of the present invention, or the concept and technical solution of the present invention are directly applied to other occasions without improvement, they are all within the protection scope of the present invention.

Claims

1. A real-time path planning method, characterized in that: The method specifically comprises the following steps: S1. Construct a topological map of the vehicle driving environment; S2, detecting whether the current position P0 of the vehicle is on a fixed path, if the detection result is no, executing step S3; S3, determine the station P1 closest to the current location on the fixed path, and plan the shortest driving path R0 from station P1 to the task station; S4, calculate the angle θ between the current position P0 and the site P1, if abs(θ)>90°, execute step S5, S5, taking the next site P2 of site P1 in path R0 as the starting point, planning the shortest path R1 from site P2 to task node Ps, and then planning the shortest driving path R2 from the current position P0 to site P2. The concatenation of path R2 and path R1 forms the driving path R from the current position P0 to task site Ps; If abs(θ)≤90°, execute step S6; S6, directly plan the shortest driving path R0 from the site P1 to the task site Ps, and then plan the shortest driving path R4 from the current position P0 to the site P1. The concatenation of the path R0 and the path R4 forms the driving path R from the current position P0 to the task site; After step S6, the method further includes: S7. After the path planning is completed, the road section weight of the current vehicle is updated, and the length of each road section in the planned path is updated based on the road section weight of the vehicle, wherein the updated road section length is equal to the sum of the initial length of the road section and the road section weight*the initial length of the road section.

2. The real-time path planning method according to claim 1, characterized in that: During the driving process of the vehicle based on the planned path, it is detected in real time whether the current position of the vehicle is the station position. If the current position is the station position, the number of vehicles on the next driving section is determined. If the number of vehicles on the next section is greater than the number threshold, the next section is marked as blocked, and the current position is used as the starting point to re-plan the shortest driving path from the current position to the task station. The vehicle drives based on the re-planned shortest driving path, and step S7 is executed for the re-planned shortest driving path.

3. The real-time path planning method according to claim 1, characterized in that: The calculation formula of the angle θ between the current position P0 and the station P1 is as follows: Among them, angle is the direction of the vehicle's current position, (x0, y0) is the coordinate of the current position P0 in the grid map, and (x1, y1) is the coordinate of the site P1 in the grid map.

4. An AGV, characterized in that: The AGV includes a processor and a memory, wherein the memory stores at least one instruction, at least one program, a code set or an instruction set, and the at least one instruction, the at least one program, the code set or the instruction set is loaded and executed by the processor to implement the real-time path planning method as described in any one of claims 1-3.

Images