Multi-AGV path planning method and device based on dynamic priority express distribution center

Abstract

The present invention relates to a multi-AGV path planning method and device based on a dynamic priority express distribution center, and belongs to the field of multi-agent control technology. The method comprises the following steps: 1. Modeling the express distribution center environment using a grid method; 2. Using an improved A* algorithm to perform initial path planning for each AGV; 3. Calculating the path time window of each AGV; 4. Finding the time and node when conflicts occur between AGVs; 5. Calculating the real-time priority of each AGV when a conflict occurs, and determining the AGV to perform collision avoidance operations; 6. Determining the type of conflict and selecting a collision avoidance method; 7. Outputting collision-free path information of each AGV. The present invention effectively solves the problem of AGV path conflicts in the express distribution center by improving the initial path planning method and designing collision avoidance rules.

Description

Technical Field

[0001] The present invention belongs to the technical field of multi-agent control, and relates to a multi-AGV path planning method and device based on a dynamic priority express distribution center. Background Art

[0002] With the rapid development of e-commerce, the express parcel business continues to expand. In 2021, the number of express parcels nationwide exceeded 108.3 billion. With the increasing number of express parcels year by year and the increasing demand of consumers for express delivery timeliness, the traditional express distribution center uses manual sorting operations to meet the actual requirements. Against this background, the automated guided vehicle (AGV) express sorting system with flexibility, efficiency and cost advantages came into being. AGV has obvious advantages. The workload of one AGV can be equivalent to 8-12 manual labors. The automated express distribution center can achieve 24-hour operation, which not only greatly reduces labor costs, but also improves operating efficiency. Reasonable path planning, solving path conflict problems, and shortening the total operating time of AGV are the key to improving AGV operating efficiency, and it is also one of the current research hotspots.

[0003] Collaborative path planning and conflict handling among multiple AGVs are the key to realizing an automated express distribution center based on multiple AGVs. The AGV path planning problem can be divided into single AGV path planning and multi-AGV path planning.

[0004] The single AGV path planning problem is essentially a problem of searching for the optimal path between the starting point and the end point. Different application scenarios have different evaluation criteria for "optimal". Common evaluation indicators include shortest time, shortest distance, least energy consumption, least number of turns, and a combination of multiple indicators. Single AGV path planning technology is divided into two categories: global path planning based on a known environment and local path planning based on sensors in an unknown environment. In terms of single AGV path planning, the focus of research is on optimizing and improving intelligent algorithms such as the A* algorithm, Dijkstra algorithm, ant colony algorithm, and genetic algorithm to obtain a better initial path. The effect of initial path planning for each AGV will greatly affect the operating efficiency of the system.

[0005] For multi-AGV path planning, the path conflict problem needs to be solved while path planning is being carried out, which is more complicated. The most important issue is that the paths between the AGVs do not conflict, followed by the optimal system path. At present, the research on multi-AGV path planning has the following problems: most studies only consider collision-free path planning under static ideal conditions, or the real-time collision avoidance methods proposed for dynamic unstable environments have poor robustness; in addition, most studies at this stage often assume that the priority of each AGV is fixed in the process of solving the AGV conflict problem, which may cause low-priority vehicles to wait for a long time when there are too many AGVs, thus causing transportation network congestion.

[0006] The document with patent number CN202110605550.6 discloses an AGV path planning method of improved A* combined with motion control, which is characterized by comprising the following steps: Step 1, according to the task requirements and the performance of the AGV mechanical structure and electronic system, design and calculate the motion control of smooth left turn and smooth right turn suitable for the task area of ​​the AGV, requiring the AGV to have a constant speed before and after turning; Step 2, according to the motion control method obtained in step 1, the minimum turning radius of the AGV in the task area is obtained, based on the minimum turning radius, the task area is rasterized, and the definition of adjacent nodes in the A* algorithm is modified in combination with the motion control method; Step 3, based on the motion control of step 1 and the raster processing of step 2, the improved A* algorithm is used to plan the path of the task area to find a smooth path from the starting point to the end point; Step 4, based on the path planned in step 3, integration processing is performed to delete redundant path points for use in motion control; Step 5, based on the path points integrated in step 4, the AGV performs motion control according to the speed mark of the path points to complete the task. The defects of this technology are that the use of the A* algorithm for initial path planning results in meaningless turns, and secondly, it does not take into account the problem of path conflicts between AGVs.

[0007] There is a need for a multi-AGV collision-free path planning method and device that can plan a better initial path and solve the AGV path conflict problem. Summary of the invention

[0008] In view of this, the purpose of the present invention is to provide a multi-AGV path planning method and device based on dynamic priority express distribution center. Aiming at the multi-AGV path planning problem of express distribution center, the present invention proposes a dynamic priority strategy combined with collision avoidance rules to solve the conflict problem of multiple AGVs. First, the A* algorithm is improved to avoid meaningless turns, and the algorithm is used to plan the initial path of each AGV to realize static path planning; then, the conflict types existing in the scene are classified and collision avoidance rules are proposed for different conflict types, and dynamic path planning is realized by quantifying the real-time priority of AGV when a conflict occurs to determine the avoidance order.

[0009] In order to achieve the above object, the present invention provides the following technical solutions:

[0010] A multi-AGV path planning method based on a dynamic priority express distribution center includes the following steps:

[0011] S1: Use the grid method to establish the environmental map of the express distribution center;

[0012] S2: Use the improved A* algorithm to perform initial path planning for each automatic guided vehicle AGV;

[0013] S3: Calculate the time window of each AGV path planned in S2;

[0014] S4: Find the time and node where the AGV conflicts through the AGV path time window calculated in S3;

[0015] S5: Calculate the real-time priority of each AGV when a conflict occurs through a dynamic priority strategy, and determine the AGV to perform collision avoidance operations;

[0016] S6: Determine the type of conflict between AGVs in S4 and select a collision avoidance method;

[0017] S7: Output collision-free AGV path information.

[0018] Optionally, in S1, the express distribution center environment map is composed of black and white grids, including the following parts:

[0019] ① Sorting station: located at the edge of the map, where AGV picks up packages;

[0020] ②Road: used for AGV driving;

[0021] ③ Package delivery area: according to the package delivery address, multiple delivery areas are scattered on the map;

[0022] ④AGV: drives in the map and carries packages;

[0023] ⑤ Package destination: The package destination corresponds to the package delivery area. The AGV needs to drive to the package delivery area corresponding to the package destination for delivery;

[0024] ⑥Obstacles: areas that occupy space and prevent AGV from passing through;

[0025] ⑦ Docking area: used for idle AGVs to dock and charge. When the current power is lower than the threshold or all tasks are completed, the AGV returns to the docking area for charging.

[0026] Optionally, in S2, the improved A* algorithm is specifically:

[0027] Based on the A* algorithm, the heuristic function is modified to avoid unnecessary turns and reduce the AGV travel time; the operation process includes the following:

[0028] S21: Obtaining the environment map information of the express distribution center, the AGV starting position S and the package location information D;

[0029] S22: define an open list for storing node information to be processed later and a close list for storing node information that has been processed, and both lists are initialized to empty lists;

[0030] S23: Put the starting point S into the list of nodes to be checked, Open list, and then find all the reachable nodes around S. The AGV can move in four directions, namely, up, down, left, and right. These are the four nodes of S. Calculate their G values, add them to the Open list, and set S as their parent node.

[0031] S24: Remove S from the Open list and add it to the checked list Close list; calculate the Manhattan distance H value from each node in the Open list to the end point, and calculate the turning cost P value; determine whether a turn occurs when the A* algorithm node is expanded. If no turn occurs, then add the G value and H value of each detection node, recorded as F; if a turn occurs, then add the G value, H value and P value of each detection node, recorded as F;

[0032] S25: Select the node with the smallest F value, check it in S21, and update the cost; if the reachable node is already in the Close list, ignore it; if not, calculate its G, H, P, F, set the parent node, and add it to the Open list; if the reachable node is already in the Open list, calculate whether moving from the current node to the node has a smaller G value, if so, reset the parent node of the node to the current node, and recalculate the G and F values;

[0033] S26: Repeat S21 to S25 until the end point D is added to the Open list to end the algorithm, and the shortest path is obtained by backtracking through the parent node.

[0034] Optionally, in S3, the path time window is specifically:

[0035] According to the initial path planned by the AGV, the running time between each AGV node is obtained by the ratio of the node length to the AGV speed.

[0036] Optionally, in S4, based on each AGV path and time window, an AGV passing through the same path node in the same time window is searched.

[0037] Optionally, in S5, the dynamic priority strategy considers factors including the current power of the AGV, whether it is currently loaded, the travel time, the remaining travel distance, the package timeliness requirement, and the path cost, and calculates the real-time priority of each AGV when a conflict occurs. The real-time priority calculation formula is:

[0038]

[0039]

[0040]

[0041]

[0042] in, is a 0-1 decision variable, which indicates the result of comparing the current power level with the threshold value when the a-th AGV transports package j. is a 0-1 decision variable, indicating whether the ath AGV is loaded. represents the timeliness requirement for the a-th AGV to transport the j-th package, U represents the maximum timeliness requirement, It indicates the running time when the ath AGV transporting package j arrives at node i. represents the total time planned for the a-th AGV to transport package j, It represents the distance that the ath AGV has traveled when it transports package j to node i. represents the total distance planned when the a-th AGV transports package j, is a 0-1 decision variable, indicating whether the a-th AGV transporting package j needs to turn when passing through node i, t vIt represents the time spent by the AGV for one turn, n represents the set of path nodes planned when the a-th AGV transports package j; α represents the weight coefficient of the timeliness requirement of the AGV transporting the package, β represents the weight coefficient of the running time of the AGV transporting the package, γ represents the weight coefficient of the distance traveled by the AGV transporting the package, and μ represents the weight coefficient of the turning cost of the AGV transporting the package. α, β, γ, μ∈(0,1), which can be changed according to the importance of each indicator by the manager; when a conflict occurs and the AGVs have the same real-time priority, the conflict is determined according to the order of the AGV numbers.

[0043] Optionally, in S6, the type of conflict between the AGVs in S4 is determined, and a collision avoidance method is selected, and there are conflicts between the AGVs in the express distribution center, such as conflicts at intersections and conflicts in occupying positions;

[0044] Opposite collision: If two AGVs are traveling in opposite directions along the same road or turning at an intersection during their forward movement, it is determined that a collision will occur. After the conflict node, the remaining paths of the two AGVs overlap with their previous paths, resulting in a mutual collision.

[0045] Intersection conflict: At an intersection, two AGVs traveling in opposite directions arrive at the same node at the same time, or two AGVs traveling in opposite directions turn at the intersection and travel in the same direction, which will cause a node conflict.

[0046] Occupancy conflict: If an AGV stays on the path of the working AGV, an occupation conflict will occur;

[0047] The collision avoidance rules are:

[0048] (1) Opposite-direction conflict: Wait in advance or re-plan the path; the AGV with lower priority waits in advance at the starting point and sets out after the AGV with higher priority passes through the repeated path, or re-plans the path for the AGV with lower task priority, and the AGV with higher task priority follows the original route;

[0049] (2) Intersection conflict: The method of ordering AGVs with lower task priority to wait for a period of time before departing from the starting point to avoid conflicts;

[0050] (3) Occupancy conflict: The node-occupying AGV is regarded as a static obstacle, and the path of the AGV with lower task priority is replanned;

[0051] The waiting strategy chooses to wait at the starting point.

[0052] Optionally, S7 specifically includes: outputting collision-free AGV path information.

[0053] A multi-AGV path planning device based on a dynamic priority express distribution center based on the method includes an express distribution center environment map construction module, an initial path planning module, a conflict detection module, a conflict resolution module and a collision-free path output module;

[0054] The express distribution center environment map building module is signal-connected to the initial path planning module;

[0055] One end of the initial path planning module is connected to the express distribution center environment map construction module by signal, and the other end is connected to the conflict detection module by signal;

[0056] One end of the conflict detection module is connected to the initial path planning module signal, and the other end is connected to the conflict resolution module signal;

[0057] One end of the conflict resolution module is connected to the conflict detection module signal, and the other end is connected to the collision-free path output module signal;

[0058] The collision-free path output module is signal-connected to the conflict resolution module;

[0059] The express distribution center environment map construction module is used to construct the AGV operating environment and generate AGV location information and package information;

[0060] The initial path planning module is used to perform initial path planning based on the AGV location information and package information in the express distribution center;

[0061] The conflict detection module calculates the time window of the path planned in the initial path planning module, and searches for the conflicting AGVs, the time and node of the conflict, and the type of conflict;

[0062] The conflict resolution module calculates the real-time priority of each AGV according to the conflicting AGVs found in the conflict detection module and the time and node of the conflict, and resolves the conflict according to the collision avoidance rules;

[0063] The collision-free path output module is used to output the collision-free path information of each AGV.

[0064] The beneficial effects of the present invention are as follows: the present invention effectively decomposes the multi-AGV path planning problem of the express distribution center into single AGV initial path planning and multi-AGV conflict resolution, improves the A* algorithm in the initial path planning stage, effectively avoids meaningless turns, and obtains a better initial path; in the conflict resolution stage, the conflict types existing in the scene are classified and collision avoidance rules are proposed for different conflict types, and the AGV path conflict problem is solved by quantifying the real-time priority of the AGV when the conflict occurs to determine the avoidance order. Experimental analysis has proved that the multi-AGV path planning method of the present invention can avoid meaningless turns, and solving the conflict problem can reduce the total task completion time and the additional time generated for solving the conflict problem compared to solving the conflict problem with a fixed priority, which can improve the operating efficiency of the express distribution center and avoid the risk of conflict caused by the long execution time of a single task.

[0065] Other advantages, objectives and features of the present invention will be described in the following description to some extent, and to some extent, will be obvious to those skilled in the art based on the following examination and study, or can be taught from the practice of the present invention. The objectives and other advantages of the present invention can be realized and obtained through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0066] In order to make the purpose, technical solutions and advantages of the present invention more clear, the present invention will be described in detail below in conjunction with the accompanying drawings, wherein:

[0067] Figure 1 It is a basic flow chart of the present invention;

[0068] Figure 2 This is a diagram of the AGV operating environment of the present invention;

[0069] Figure 3 This is a schematic diagram of the improved A* algorithm flow of the present invention;

[0070] Figure 4 A schematic diagram of collision avoidance types is provided for the present invention;

[0071] Figure 5 It is a schematic diagram of the device of the present invention. DETAILED DESCRIPTION

[0072] The following describes the embodiments of the present invention by specific examples, and those skilled in the art can easily understand other advantages and effects of the present invention from the contents disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that the illustrations provided in the following embodiments only illustrate the basic concept of the present invention in a schematic manner, and the following embodiments and features in the embodiments can be combined with each other without conflict.

[0073] Among them, the drawings are only used for illustrative explanations, and they only represent schematic diagrams rather than actual pictures, and should not be understood as limitations on the present invention. In order to better illustrate the embodiments of the present invention, some parts of the drawings may be omitted, enlarged or reduced, and do not represent the size of actual products. For those skilled in the art, it is understandable that some well-known structures and their descriptions in the drawings may be omitted.

[0074] The same or similar numbers in the drawings of the embodiments of the present invention correspond to the same or similar parts; in the description of the present invention, it should be understood that if the terms "upper", "lower", "left", "right", "front", "rear", etc. indicate the orientation or position relationship, they are based on the orientation or position relationship shown in the drawings, which is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operate in a specific orientation. Therefore, the terms describing the position relationship in the drawings are only used for illustrative purposes and cannot be understood as limiting the present invention. For ordinary technicians in this field, the specific meanings of the above terms can be understood according to specific circumstances.

[0075] See also Figure 1 to Figure 5 , which is a multi-AGV path planning method and device for express distribution center based on dynamic priority strategy.

[0076] Example 1: Figure 1 A method and device for multi-AGV path planning in an express distribution center based on a dynamic priority strategy is shown, including but not limited to the following steps:

[0077] Step 1: Enter the express distribution center information to build the environment map. Figure 2As shown in the figure, the environmental map of the express distribution center in this method is composed of black and white grids, including the following parts: ① Sorting station: located at the edge of the map, where the AGV picks up the package; ② Road: used for AGV driving; ③ Package delivery area: according to the delivery address of the package, multiple delivery areas are scattered on the map; ④ AGV: driving in the map and carrying the package; ⑤ Package destination: the package destination corresponds to the package delivery area, and the AGV needs to drive to the package delivery area corresponding to the package destination for delivery; ⑥ Obstacle: an area that occupies space and makes it impossible for the AGV to pass through; ⑦ Docking area: used for the docking and charging of idle AGVs. When the current power is lower than the threshold or all tasks are completed, the AGV returns to the docking area for charging.

[0078] Step 2: Use the improved A* algorithm to perform initial path planning for each AGV. The details are as follows: Based on the information environment map of the express distribution center, use the improved A* algorithm to find the optimal initial path for each AGV in the environment. The initial path is the path from the current position of the AGV to the sorting station to pick up the task or from the sorting station to the destination of the package after picking up the task.

[0079] Step 3: Calculate the time window of each AGV path. Specifically, the improved A* algorithm is used to find the optimal initial path for each AGV in the environment, and the time it takes for each AGV to pass through the path node is calculated through the path, speed, and additional turning time.

[0080] Step 4: Find the time and node where the conflict occurs between AGVs. Specifically, based on the time window information of each AGV path, find the AGV information that passes through the same path node in the same time window.

[0081] Step 5: Calculate the real-time priority of the conflicting AGVs according to the dynamic priority strategy. Specifically, calculate the real-time priority of each AGV according to the dynamic priority calculation formula based on the found conflicting AGVs.

[0082] Step 6: Determine the type of collision between AGVs and select a collision avoidance method. Specifically, according to the actual operation of the express distribution center, determine the three types of conflicts: opposite-direction conflict, intersection conflict, and node occupation, and set a waiting or re-path planning collision avoidance method for each type.

[0083] Step 7: Output collision-free AGV path information.

[0084] Application examples: See Figure 1 , a method and device for multi-AGV path planning in an express distribution center based on a dynamic priority strategy, comprising the following steps:

[0085] Step 1: Abstract the actual environment of the express distribution center to obtain the environment in which the AGV operates. According to the characteristics of the express distribution center, the grid method is selected for environmental modeling. The final environment is as follows: Figure 2 As shown. In this method, the environment map of the express distribution center is composed of black and white grids, including the following parts: ① Sorting station: located at the edge of the map, where AGV picks up the package; ② Road: used for AGV driving; ③ Package delivery area: according to the delivery address of the package, multiple delivery areas are scattered on the map; ④ AGV: driving in the map and carrying the package; ⑤ Package destination: the package destination corresponds to the package delivery area, and the AGV needs to drive to the package delivery area corresponding to the package destination for delivery; ⑥ Obstacles: areas that occupy space and make it impossible for AGV to pass; ⑦ Docking area: used for idle AGVs to dock and charge. When the current power is lower than the threshold or all tasks are completed, they return to the docking area for charging.

[0086] Step 2: Use the improved A* algorithm to plan the path for each AGV and obtain the optimal initial path under ideal conditions. Figure 3 Flowchart for planning the initial path of AGV using the improved A* algorithm.

[0087] Taking the following four AGV tasks in Table 1 as an example, the path information planned using the improved A* algorithm is shown in Table 2.

[0088] Table 1 Four AGV tasks

[0089]

[0090] To simplify the expression, the coordinates are converted into index values. The specific forwarding method is to start from the origin (0,0), the grid index value starts from 1, and proceeds from left to right and from bottom to top. Assuming that the two-dimensional number of the grid is (x, y), the corresponding formula between the two-dimensional coordinates and the index value is: Index value = (x+0.5)+(y+0.5-1)*22

[0091] Table 2 Path information planned using the improved A* algorithm

[0092]

[0093]

[0094] Step 3: Calculate the time window of each AGV. The calculation results are shown in Table 3.

[0095] Table 3 Calculation results of each AGV time window

[0096]

[0097] Step 4: Find the time and node where the conflict occurs between AGVs. Through the above AGV path time window information, it is calculated that AGV1 and AGV3 collide at node 218 at the 24th second, AGV2 and AGV4 collide at node 117 at the 22nd second, and AGV3 and AGV4 collide at node 264 at the 8th second.

[0098] Step 5. Use the dynamic priority strategy to calculate the real-time priority of the AGV when it collides. In this embodiment, α=β=γ=μ=0.25 is taken, and the power threshold is set to 10%. AGV3 and AGV4 collide at node 264, and the real-time priority of the two AGVs at the collision node is calculated using the formula, where p4=100.191, p3=10.465; AGV1 and AGV3 collide at node 219, and the real-time priority of the two AGVs at the collision node is calculated using the formula, where P1=10.535, p3=10.562; AGV1 and AGV2 collide at node 306, and the current priority of the two AGVs is calculated using the formula, where p1=10.652, p2=0.517.

[0099] Step 6, determine the conflict type and select the collision avoidance method. AGV2 and AGV4 have a placeholder conflict at node 117. The real-time priority of the two AGVs at the collision node is calculated using the formula, where P2 = 10.535, p4 = 10.562. AGV2 re-plans the path as follows: 47-48-49-50-51-52-73-94-115-136-137-138-159-180-201-222-243-264-285-306-305. AGV1 and AGV3 have a conflict at node 219. The real-time priority of the two AGVs at the collision node is calculated using the formula, where P1 = 10.535, p3 = 10.562. Because AGV3 has a higher real-time priority at this node, AGV3 maintains the original plan and AGV1 makes adjustments. AGV1 considers two measures: waiting at the starting point to let AGV3 pass the conflict node first and re-planning the path. AGV1 needs to wait for 26 seconds at the starting point to avoid the conflict with AGV3. After avoiding the conflict, the total time for AGV1 to complete the task is 74 seconds. The AGV re-plans the path. The re-planned path is: 212-233-254-255-256-257-258-279-300-321-342-343-344-345-346-347-348-327-306-285-264-265-266-267. After avoiding the conflict, the total time for AGV1 to complete the task is 56 seconds.

[0100] AGV1 and AGV2 collide at node 306. The current priorities of the two AGVs are calculated using the formula, where p1 = 10.652 and p2 = 0.517. Since AGV1 has a higher real-time priority, AGV1 keeps moving forward while AGV2 makes adjustments. AGV2 considers two measures: waiting at the starting point to let AGV1 pass the conflicting node first and replanning the path. AGV2 needs to wait for 8 seconds at the starting point to avoid conflict with AGV1. After avoiding the conflict, the total time for AGV2 to complete the task is 56 seconds. The path of AGV2 is replanned. The replanned path is: 47-46-67-88-109-130-151-172-171-170-191-212-233-254-255-256-257-258-279-300-321-342-343-344-345-346-325-304-305. After avoiding the conflict, the total time for AGV2 to complete the task is 72 seconds.

[0101] Step 7: Output collision-free AGV path information.

[0102] Figure 4 A schematic diagram of collision avoidance types is provided for the present invention;

[0103] Figure 5 It is a schematic diagram of the device of the present invention.

[0104] An embodiment of the present invention provides an AGV path planning device, comprising:

[0105] Environmental map construction module: used to build the environment in which AGV operates.

[0106] In a preferred embodiment, the express distribution center environment map construction module includes an input device and a memory, wherein the input device is used to input the layout information of the express distribution center, and the memory is used to store the AGV driving environment.

[0107] Initial path planning module: used to plan the initial path for the AGV according to the AGV driving environment.

[0108] In a preferred embodiment, the initial path planning module includes an input device, a first processor unit, and an output device. The input device is used to input the location information and destination information of the AGV, the first processor unit is used to execute a computer program that uses the improved A* algorithm to plan an initial path for the AGV, and the output device is used to output and display the initial path obtained by the first processor unit.

[0109] Conflict detection module: calculates the time window and determines the conflict type for all planned AGV paths.

[0110] In a preferred embodiment, the conflict detection module includes an input device and a second processor unit. The input device is used to input the planned AGV initial path, and the second processor unit is used to determine whether a path conflict occurs between AGVs and the type of conflict.

[0111] Conflict resolution module: Use dynamic priority strategy to calculate the real-time priority of conflicting AGVs and select collision avoidance method to eliminate conflicts.

[0112] In a preferred embodiment, the conflict resolution module includes an output device and a processor unit 3. The output device is used to input necessary information for calculating the real-time priority of the AGV, and the processor unit 3 is used to calculate the real-time priority of the conflicting AGV according to the dynamic priority formula, so as to select the AGV for collision avoidance operation.

[0113] Collision-free path output module: outputs collision-free path information for each AGV.

[0114] In a preferred embodiment, the collision-free path output module includes a fourth processor unit. The fourth processor unit is used to output path information showing that each AGV has no collisions.

[0115] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solution of the present invention can be modified or replaced by equivalents without departing from the purpose and scope of the technical solution, which should be included in the scope of the claims of the present invention.

Claims

1. A multi-AGV path planning method based on dynamic priority express distribution center, characterized by: The method comprises the following steps: S1: Use the grid method to establish the environmental map of the express distribution center; S2: Use the improved A* algorithm to perform initial path planning for each automatic guided vehicle AGV; S3: Calculate the time window of each AGV path planned in S2; S4: Find the time and node where the AGV conflicts through the AGV path time window calculated in S3; S5: Calculate the real-time priority of each AGV when a conflict occurs through a dynamic priority strategy, and determine the AGV to perform collision avoidance operations; S6: Determine the type of conflict between AGVs in S4 and select a collision avoidance method; S7: Output collision-free AGV path information; In S1, the environment map of the express distribution center is composed of black and white grids, including the following parts: ① Sorting station: located at the edge of the map, where AGV picks up packages; ②Road: used for AGV driving; ③ Package delivery area: according to the package delivery address, multiple delivery areas are scattered on the map; ④AGV: drives in the map and carries packages; ⑤ Package destination: The package destination corresponds to the package delivery area. The AGV needs to drive to the package delivery area corresponding to the package destination for delivery; ⑥Obstacles: areas that occupy space and prevent AGV from passing through; ⑦ Docking area: used for idle AGVs to dock and charge. When the current power is lower than the threshold or all tasks are completed, the AGV returns to the docking area for charging. In S2, the improved A* algorithm is specifically: Based on the A* algorithm, the heuristic function is modified to avoid unnecessary turns and reduce the AGV travel time; the operation process includes the following: S21: Obtaining the environment map information of the express distribution center, the AGV starting position S and the package location information D; S22: define an open list for storing node information to be processed later and a close list for storing node information that has been processed, and both lists are initialized to empty lists; S23: Put the starting point S into the list of nodes to be checked, Open list, and then find all the reachable nodes around S. The AGV can move in four directions, namely, up, down, left, and right. These are the four nodes of S. Calculate their G values, add them to the Open list, and set S as their parent node. S24: Remove S from the Open list and add it to the checked list Close list; calculate the Manhattan distance H value from each node in the Open list to the end point, and calculate the turning cost P value; determine whether a turn occurs when the A* algorithm node is expanded. If no turn occurs, then add the G value and H value of each detection node, recorded as F; if a turn occurs, then add the G value, H value and P value of each detection node, recorded as F; S25: Select the node with the smallest F value, check it in S21, and update the cost; if the reachable node is already in the Close list, ignore it; if not, calculate its G, H, P, F, set the parent node, and add it to the Open list; if the reachable node is already in the Open list, calculate whether moving from the current node to the node has a smaller G value, if so, reset the parent node of the node to the current node, and recalculate the G and F values; S26: Repeat S21 to S25 until the end point D is added to the Open list to end the algorithm, and the shortest path is obtained by backtracking through the parent node; In S3, the path time window is specifically: According to the initial path planned by AGV, the running time between each AGV node is obtained by the ratio of node length to AGV speed; In S4, according to each AGV path and time window, find the AGV passing through the same path node in the same time window; In S5, the dynamic priority strategy considers factors including the current power of the AGV, whether it is currently loaded, the travel time, the remaining travel distance, the package timeliness requirement, and the path cost, and calculates the real-time priority of each AGV when a conflict occurs. The real-time priority calculation formula is: in, is a 0-1 decision variable, which indicates the result of comparing the current power level with the threshold value when the a-th AGV transports package j. is a 0-1 decision variable, indicating whether the ath AGV is loaded. represents the timeliness requirement for the a-th AGV to transport the j-th package, U represents the maximum timeliness requirement, It indicates the time that the ath AGV transporting package j has been running when it arrives at node i. represents the total time planned for the a-th AGV to transport package j, It represents the distance that the ath AGV has traveled when it transports package j to node i. represents the total distance planned when the a-th AGV transports package j, is a 0-1 decision variable, indicating whether the a-th AGV transporting package j needs to turn when passing through node i, t v It represents the time spent by the AGV on a turn, n represents the set of nodes planned for the path when the a-th AGV transports package j; α represents the weight coefficient of the timeliness requirement of the AGV transporting the package, β represents the weight coefficient of the running time of the AGV transporting the package, γ represents the weight coefficient of the distance traveled by the AGV transporting the package, and μ represents the weight coefficient of the turning cost of the AGV transporting the package. α, β, γ, μ∈(0,1), which changes according to the importance of each indicator according to the manager; when a conflict occurs and the AGVs have the same real-time priority, the order of the AGV numbers is used to determine the conflict.

2. The multi-AGV path planning method based on a dynamic priority express distribution center according to claim 1 is characterized by: In S6, the type of conflict between AGVs in S4 is determined, and a collision avoidance method is selected. There are conflicting directions, intersection conflicts, and position conflicts among AGVs in the express distribution center. Opposite collision: If two AGVs are traveling in opposite directions along the same road or turning at an intersection during their forward movement, it is determined that a collision will occur. After the conflict node, the remaining paths of the two AGVs overlap with their previous paths, resulting in a mutual collision. Intersection conflict: At an intersection, two AGVs traveling in intersecting directions arrive at the same node at the same time, or two AGVs traveling in opposite directions turn at the intersection and travel in the same direction, which will cause a node conflict; Occupancy conflict: If an AGV stays on the path of the working AGV, an occupation conflict will occur; The collision avoidance rules are: (1) Opposite-direction conflict: Wait in advance or re-plan the path; the AGV with lower priority waits in advance at the starting point and sets out after the AGV with higher priority passes through the repeated path, or re-plans the path for the AGV with lower task priority, and the AGV with higher task priority follows the original route; (2) Intersection conflict: The method of ordering AGVs with lower task priority to wait for a period of time before departing from the starting point to avoid conflicts; (3) Occupancy conflict: The node-occupying AGV is regarded as a static obstacle, and the path of the AGV with lower task priority is replanned; The waiting strategy chooses to wait at the starting point.

3. The multi-AGV path planning method based on a dynamic priority express distribution center according to claim 2 is characterized in that: The S7 specifically includes: outputting collision-free AGV path information.

4. A multi-AGV path planning device based on a dynamic priority express distribution center based on the method of any one of claims 1 to 3, characterized in that: The device includes an express distribution center environment map construction module, an initial path planning module, a conflict detection module, a conflict resolution module and a collision-free path output module; The express distribution center environment map building module is signal-connected to the initial path planning module; One end of the initial path planning module is connected to the express distribution center environment map construction module by signal, and the other end is connected to the conflict detection module by signal; One end of the conflict detection module is connected to the initial path planning module signal, and the other end is connected to the conflict resolution module signal; One end of the conflict resolution module is connected to the conflict detection module signal, and the other end is connected to the collision-free path output module signal; The collision-free path output module is signal-connected to the conflict resolution module; The express distribution center environment map construction module is used to construct the AGV operating environment and generate AGV location information and package information; The initial path planning module is used to perform initial path planning based on the AGV location information and package information in the express distribution center; The conflict detection module calculates the time window of the path planned in the initial path planning module, and searches for the conflicting AGVs, the time and node of the conflict, and the type of conflict; The conflict resolution module calculates the real-time priority of each AGV according to the conflicting AGVs found in the conflict detection module and the time and node of the conflict, and resolves the conflict according to the collision avoidance rules; The collision-free path output module is used to output the collision-free path information of each AGV.

Images