Multi-intelligent carrying AGV conflict coordination and dynamic obstacle avoidance cooperation path planning method

Abstract

The invention provides a multi-intelligent carrying AGV conflict coordination and dynamic obstacle avoidance cooperation path planning method, and relates to the field of transportation scheduling, and the method comprises the steps: obtaining environment information, building a grid map, and setting starting point nodes and target nodes of a plurality of intelligent carrying AGVs; determining a global optimal path of each intelligent carrying AGV, and judging whether a conflict occurs or not; if the conflict is determined to be generated, based on the global optimal path of each intelligent carrying AGV, determining conflict types of space-time conflicts and time-space conflicts on the global optimal paths of the plurality of intelligent carrying AGVs; for each time-space conflict, determining the priority of the intelligent carrying AGV with the time-space conflict, and determining a conflict resolution strategy corresponding to the time-space conflict based on the priority of the intelligent carrying AGV with the time-space conflict and the conflict type of the time-space conflict; and based on the global optimal path of each intelligent carrying AGV and the conflict resolution strategy corresponding to each time-space conflict, a plurality of intelligent carrying AGVs are scheduled, and the method has the advantage of realizing collaborative carrying in a multi-carrying AGV scene.

Description

Technical Field

[0001] The present invention relates to the field of transport scheduling, and in particular to a method for conflict coordination and dynamic obstacle avoidance collaborative path planning for multiple intelligent transport AGVs. Background Art

[0002] With the advancement of technology, automated guided vehicles (AGVs) are increasingly taking on the handling and picking tasks in warehouses. However, as the scale of factory production continues to expand and production lines become more complex, the problem-solving capabilities of a single handling AGV are limited. For many complex tasks, multiple handling AGVs are required to work together to complete the task.

[0003] In the prior art, a distributed AGV path planning method is usually used to plan the path for AGVs. However, when performing AGV collaborative path planning, there are many redundant points, low operating efficiency, poor path quality, inability to dynamically avoid obstacles, and weak real-time conflict coordination.

[0004] Therefore, it is necessary to provide a multi-intelligent transport AGV conflict coordination and dynamic obstacle avoidance collaborative path planning method to solve the path planning problem of multi-intelligent transport AGVs in an unknown environment with dynamic obstacles. Summary of the invention

[0005] The present invention provides a method for conflict coordination and dynamic obstacle avoidance collaborative path planning of multiple intelligent transport AGVs, comprising: acquiring environmental information, establishing a grid map, and setting the starting node and target node of multiple intelligent transport AGVs; determining the global optimal path of each of the intelligent transport AGVs; judging whether a conflict occurs based on the global optimal path of each of the intelligent transport AGVs; if it is determined that a conflict occurs, generating a multi-AGV path conflict tree based on the global optimal path of each of the intelligent transport AGVs, wherein the multi-AGV path conflict tree is used to record the spatiotemporal conflicts on the global optimal paths of multiple intelligent transport AGVs and the conflict types of the spatiotemporal conflicts; for each of the spatiotemporal conflicts, determining the priority of the intelligent transport AGV that has the spatiotemporal conflict, and determining the conflict resolution strategy corresponding to the spatiotemporal conflict based on the priority of the intelligent transport AGV that has the spatiotemporal conflict and the conflict type of the spatiotemporal conflict; scheduling the multiple intelligent transport AGVs based on the global optimal path of each of the intelligent transport AGVs and the conflict resolution strategy corresponding to each of the spatiotemporal conflicts.

[0006] Furthermore, the global optimal path of each of the intelligent transport AGVs is determined, including: S11, establishing an open list and a closed list, and adding the starting node of the intelligent transport AGV to the open list, wherein the open list is used to record all nodes that have been generated but not yet fully explored, and the closed list is used to record all nodes that have been explored; S12, introducing the probability of environmental obstacles, adjusting the weight of the heuristic function according to the probability of environmental obstacles, and establishing an adaptive evaluation function; S13, based on the adaptive evaluation function, calculating the evaluation function value of each node in the open list, and determining the visited node from the open list, wherein the visited node is the node with the smallest evaluation function value in the open list; S14, for each adjacent node of the visited node, if the adjacent node is not recorded in the closed list and is not If the adjacent node is recorded in the open list, the adjacent node is added to the open list. If the adjacent node is recorded in the closed list, the adjacent node is ignored. If the adjacent node is not recorded in the closed list but is recorded in the open list, the path length between the access node and the adjacent node is calculated. If the path length between the access node and the adjacent node is shorter than the path length between the parent node of the adjacent node and the adjacent node, the parent node of the adjacent node is updated to the access node; S15, determine whether the access node is the target node of the intelligent transport AGV or whether the open list is empty. If not, execute S13. If so, generate the current global path; S16, introduce a redundant point deletion strategy to perform secondary optimization on the current global path to generate the global optimal path of the intelligent transport AGV.

[0007] Furthermore, the adaptive evaluation function is: ,in, For the comprehensive cost, is the actual cost from the starting node to the current node, is the distance between the current node and the target node, is the distance from the starting node to the target node, is the probability of environmental obstacles, is the estimated cost from the current node to the target node.

[0008] Furthermore, the redundant point deletion strategy is introduced to perform secondary optimization on the current global path to generate the global optimal path of the intelligent handling AGV, including: S21, taking the next node of the starting point node in the current global path as the current node; S22, judging whether the current node is on the same straight line as the previous node and the parent node of the previous node and has not passed through obstacles. If the current node, the previous node and the parent node of the previous node are on the same straight line and have not passed through obstacles, then other nodes on the straight line are regarded as redundant nodes, and the redundant nodes are deleted from the current global path. If the current node, the previous node and the parent node of the previous node are not on the same straight line or there is an obstacle on the straight line, then the previous node is regarded as a key turning point; S23, judging whether all nodes of the current global path are traversed, if not, executing S24, if yes, generating the global optimal path of the intelligent handling AGV; S24, taking the next node of the current node as the current node, and executing S22.

[0009] Furthermore, determining the priority of the intelligent transport AGV having the time-space conflict includes: determining the priority of the intelligent transport AGV having the time-space conflict based on the task energy consumption, task completion and load status of the intelligent transport AGV having the time-space conflict.

[0010] Furthermore, based on the priority of the intelligent transport AGV that has the time-space conflict and the conflict type of the time-space conflict, a conflict resolution strategy corresponding to the time-space conflict is determined, including: when the conflict type of the time-space conflict is a node conflict, based on the priority of the intelligent transport AGV that has the time-space conflict, a waiting intelligent transport AGV is determined, and the waiting intelligent transport AGV temporarily waits until the node conflict ends, wherein the node conflict is that at least two of the intelligent transport AGVs will occupy the same node together at the next moment and travel in different directions; when the conflict type of the time-space conflict is a head-on conflict, based on the intelligent transport AGV that has the time-space conflict The priority of the intelligent transport AGV to be avoided is determined, the avoidance strategy of the intelligent transport AGV to be avoided is determined, and the intelligent transport AGV to be avoided is scheduled to execute the avoidance strategy, wherein the opposite conflict means that at least two of the intelligent transport AGVs will jointly occupy each other's paths at the next moment and travel in opposite directions; when the conflict type of the time-space conflict is a same-direction conflict, the waiting intelligent transport AGV is determined based on the priority of the intelligent transport AGV in which the time-space conflict occurs, and the waiting intelligent transport AGV temporarily waits until the same-direction conflict ends, wherein the same-direction conflict means that at least two of the intelligent transport AGVs travel in the same direction after the conflict occurs.

[0011] Furthermore, based on the global optimal path of each of the intelligent transport AGVs and the conflict resolution strategy corresponding to each of the space-time conflicts, the multiple intelligent transport AGVs are scheduled, including: if the intelligent transport AGV encounters an unknown dynamic obstacle during driving, the speed of the intelligent transport AGV is adaptively adjusted to perform dynamic obstacle avoidance; if the intelligent transport AGV encounters a space-time conflict during driving, based on the global optimal path of each of the intelligent transport AGVs and the conflict resolution strategy corresponding to the space-time conflict, the multiple intelligent transport AGVs are scheduled.

[0012] Furthermore, the adaptively adjusting the speed of the intelligent transport AGV according to the complexity of the environment includes: adaptively adjusting the speed of the intelligent transport AGV based on an improved dynamic window method.

[0013] Furthermore, the speed of the intelligent transport AGV is adaptively adjusted based on the improved dynamic window method, including: sampling to obtain a speed combination; simulating multiple motion trajectories based on the sampled speed combination; evaluating each of the motion trajectories through the evaluation function of the improved dynamic window method to determine the optimal motion trajectory; and adjusting the speed of the intelligent transport AGV based on the optimal motion trajectory.

[0014] Furthermore, the evaluation function of the improved dynamic window method is: , , in, is the evaluation score of the motion trajectory, , , and are weight coefficients, is the smoothing coefficient, is the angular deviation between the end point direction of the simulated motion trajectory at the current speed and the target node, is the current estimated running speed, is the path curvature parameter, is the minimum straight-line distance between the end point of the simulated motion trajectory and the known obstacle, is the minimum straight-line distance between the end point of the simulated motion trajectory and the unknown obstacle, For the current critical turning point Axis coordinates, For the current critical turning point Axis coordinates, For the target node Axis coordinates, For the target node Axis coordinates, is the pose angle of the line connecting the current key turning point and the target node in the global coordinate system, The current sub-target point Axis coordinates, For the next sub-target point Axis coordinates, The current sub-target point Axis coordinates, For the next sub-target point Axis coordinates.

[0015] Compared with the prior art, the method for multi-intelligent handling AGV conflict coordination and dynamic obstacle avoidance collaborative path planning provided by the present invention has at least the following beneficial effects: 1. Introduce the probability of environmental obstacles, adjust the weight of the heuristic function according to the probability of environmental obstacles, establish an adaptive evaluation function, and adjust the weight of the heuristic function in an adaptive manner; introduce a redundant point deletion strategy to eliminate redundant nodes and reduce traversal nodes and inflection points, making the path smoother, improving the smoothness and low safety of the path planned by the A* algorithm, and not meeting the AGV motion characteristics. The problems improve the efficiency and accuracy of path planning; 2. On the basis of obtaining the single AGV planned path, a multi-AGV path conflict tree is established to determine whether conflicts occur within the multi-intelligent handling AGV system and the types of conflicts that may occur; considering the energy consumption of each AGV, a conflict resolution strategy based on task priority is used to coordinate conflicts between AGVs. In a complex environment considering AGV conflicts, multi-AGV random obstacle avoidance can be achieved, and the planned path has good security, real-time and universality; 3. Based on the improved dynamic window method, the speed of the intelligent transport AGV is adaptively adjusted, so that the AGV can avoid random dynamic obstacles in the environment. This improves the existing fusion algorithm's tendency to fall into local optimality and the equal treatment of conflict distances of dynamic obstacles and static obstacles. The planned path is smoother and safer, and more in line with the motion characteristics of the intelligent transport AGV, so that the intelligent transport AGV can smoothly reach the target node while improving safety. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] This specification will be further described in the form of exemplary embodiments, which will be described in detail by the accompanying drawings. These embodiments are not restrictive, and in these embodiments, the same number represents the same structure, wherein: Figure 1 It is a flowchart of a method for multi-intelligent transport AGV conflict coordination and dynamic obstacle avoidance collaborative path planning according to some embodiments of this specification; Figure 2is a schematic diagram of node conflicts according to some embodiments of this specification; Figure 3 is a schematic diagram of a conflicting direction according to some embodiments of the present specification; Figure 4 is a schematic diagram of same-direction conflicts according to some embodiments of the present specification. DETAILED DESCRIPTION

[0017] In order to more clearly illustrate the technical solutions of the embodiments of this specification, the following is a brief introduction to the drawings required for the description of the embodiments. Obviously, the drawings described below are only some examples or embodiments of this specification. For ordinary technicians in this field, this specification can also be applied to other similar scenarios based on these drawings without creative work. Unless it is obvious from the language environment or otherwise explained, the same reference numerals in the figures represent the same structure or operation.

[0018] Figure 1 is a flow chart of a method for multi-intelligent transport AGV conflict coordination and dynamic obstacle avoidance collaborative path planning according to some embodiments of this specification, such as Figure 1 As shown, a method for conflict coordination and dynamic obstacle avoidance collaborative path planning of multiple intelligent transport AGVs may include the following steps.

[0019] Step 110, obtain environmental information, establish a grid map, and set the starting nodes and target nodes of multiple intelligent transport AGVs.

[0020] Specifically, the environmental information can be obtained through the sensors of the intelligent handling AGV (for example, laser radar (LiDAR), camera, infrared sensor, ultrasonic sensor, etc.) to establish a grid map. For example, after obtaining the sensor data, a series of data processing is required, including data cleaning, feature extraction, data fusion and other steps to generate effective information that can be used for grid map construction and establish a grid map. A grid map is a map representation method that divides the environment into a series of grids, each grid represents an area in the environment and contains certain information about the area (for example, whether there are obstacles).

[0021] Step 120, determining the global optimal path for each intelligent transport AGV.

[0022] Specifically include: S11, establishing an open list and a closed list, and adding the starting node of the intelligent handling AGV to the open list, wherein the open list is used to record all nodes that have been generated but not fully explored, and the closed list is used to record all nodes that have been explored; S12, introducing the probability of environmental obstacles, adjusting the weight of the heuristic function according to the probability of environmental obstacles, and establishing an adaptive evaluation function; S13, based on the adaptive evaluation function, calculating the evaluation function value of each node in the open list, and determining the access node from the open list, wherein the access node is the node with the smallest evaluation function value in the open list; S14, for each adjacent node of the visited node, if the adjacent node is not recorded in the closed list and is not recorded in the open list, the adjacent node is added to the open list; if the adjacent node is recorded in the closed list, the adjacent node is ignored; if the adjacent node is not recorded in the closed list and is recorded in the open list, the path length between the visited node and the adjacent node is calculated; if the path length between the visited node and the adjacent node is shorter than the path length between the parent node of the adjacent node and the adjacent node, the parent node of the adjacent node is updated to be the visited node; S15, determine whether the access node is the target node of the intelligent handling AGV or whether the open list is empty. If not, execute S13. If yes, generate the current global path. Specifically, the path can be obtained by backtracking the parent node to obtain the current global path. If the open list is empty, it means that there is no path from the starting node to the target node. S16. Introduce a redundant point deletion strategy to perform secondary optimization on the current global path to generate the global optimal path of the intelligent transport AGV.

[0023] In some embodiments, the adaptive merit function is: , , , , , in, For the comprehensive cost, is the actual cost from the starting node to the current node, is the distance between the current node and the target node, is the distance from the starting node to the target node, is the probability of environmental obstacles, is the estimated cost from the current node to the target node, , , For the current node Axis coordinates, For the current node Axis coordinates, For the target node Axis coordinates, For the target node Axis coordinates, For the starting node Axis coordinates, For the starting node Axis coordinates.

[0024] It can be calculated by summing up the path costs from the starting point to the current node. For example, if the cost of each move is 1, then It is the number of moves on the path from the starting node to the current node.

[0025] The weight of the heuristic function is adjusted according to the probability of environmental obstacles. When the current node is far away from the target point, is much smaller than the actual cost, which improves the search efficiency; when the current node is close to the target point, Close to the actual cost, reduce the search speed; when there are few obstacles between the current point and the target point, increase to improve the search efficiency; when there are many obstacles between the current point and the target point, reduce to avoid the search path falling into the local optimum.

[0026] In some embodiments, a redundant point deletion strategy is introduced to perform secondary optimization on the current global path to generate a global optimal path for the intelligent transport AGV, including: S21, taking the next node of the starting point node in the current global path as the current node; S22, determine whether the current node is on the same straight line as the previous node and the parent node of the previous node and has not passed through an obstacle. If the current node, the previous node and the parent node of the previous node are on the same straight line and have not passed through an obstacle, then other nodes on the straight line are regarded as redundant nodes, and the redundant nodes are deleted from the current global path. If the current node, the previous node and the parent node of the previous node are not on the same straight line or there is an obstacle on the straight line, the previous node is regarded as a key turning point. If the current node, the previous node and the parent node of the previous node are not on the same straight line or there is an obstacle on the straight line, then the previous node is a must-pass node, because the path must change direction or bypass an obstacle here, so the previous node is a key turning point on the path; S23, determining whether all nodes of the current global path are traversed, if not, executing S24, if yes, generating a global optimal path of the intelligent handling AGV, wherein the global optimal path of the intelligent handling AGV only includes the path of the starting point, the key turning point and the end point, so that the planned global optimal path meets the motion characteristics of the intelligent handling AGV; S24. Take the next node of the current node as the current node and execute S22.

[0027] Step 130 , based on the global optimal path of each intelligent transport AGV, determine whether a conflict occurs.

[0028] Depending on whether the position of the AGV and the object causing the conflict has changed, it can be divided into static conflict and dynamic conflict. Static conflict refers to the conflict between the intelligent handling AGV and the known static obstacles in the environment during operation; dynamic conflict refers to the conflict between the intelligent handling AGV and the moving object at the same position, including AGV internal conflict and AGV dynamic conflict. According to the different AGV conflict intervals and paths, multi-AGV internal conflicts can be divided into three situations: node conflict, opposite conflict and same-direction conflict.

[0029] Node conflict means that at least two intelligent handling AGVs will occupy the same node at the next moment and travel in different directions. Figure 2 is a schematic diagram of node conflicts according to some embodiments of this specification, such as Figure 2 As shown, at the intersection, the intelligent handling AGV1 and the intelligent handling AGV2 will occupy the same node together at the next moment and travel in different directions.

[0030] A conflict in the opposite direction means that at least two intelligent handling AGVs will occupy each other's paths at the next moment and drive in opposite directions. Figure 3 is a schematic diagram of the conflicting directions shown in some embodiments of this specification, such as Figure 3 As shown, in the same driving channel, the intelligent transport AGV 1 and the intelligent transport AGV 2 occupy each other's path and drive in opposite directions.

[0031] The same-direction conflict occurs when at least two intelligent handling AGVs travel in the same direction after a conflict occurs. Figure 4 is a schematic diagram of a same-direction conflict according to some embodiments of this specification, such as Figure 4 As shown, in the same driving channel, the intelligent transport AGV1 and the intelligent transport AGV2 continue to travel in the same direction after a conflict occurs at a node.

[0032] The dynamic conflict of intelligent handling AGV refers to the unknown dynamic obstacles that exist in the path during the operation of each intelligent handling AGV.

[0033] Step 140 , if it is determined that a conflict occurs, a multi-AGV path conflict tree is generated based on the global optimal path of each intelligent transport AGV, wherein the multi-AGV path conflict tree is used to record the spatiotemporal conflicts and the conflict types of the spatiotemporal conflicts on the global optimal paths of multiple intelligent transport AGVs.

[0034] Step 150 , for each space-time conflict, determine the priority of the smart transport AGV that has the space-time conflict, and determine the conflict resolution strategy corresponding to the space-time conflict based on the priority of the smart transport AGV that has the space-time conflict and the conflict type of the space-time conflict.

[0035] In some embodiments, the task energy consumption, task completion degree and load status of the smart transport AGV having the time-space conflict determine the priority of the smart transport AGV having the time-space conflict.

[0036] The task energy consumption is used to measure the difficulty of the intelligent handling AGV path, that is, the ratio of the number of turns in each intelligent handling AGV path to the total length of the path. Since the intelligent handling AGV has a turning cost, the higher the task energy consumption, the more difficult it is to complete the task, and the higher the priority is set accordingly. The formula is as follows: , In the formula, For the Energy consumption of each intelligent handling AGV task; For the The number of turns in the global optimal path of an intelligent handling AGV; For the The length of the global optimal path of an intelligent handling AGV.

[0037] The task completion rate indicates the progress of the intelligent handling AGV in completing the task, which is the ratio of the traveled distance to the total distance. The larger the ratio, the higher the task completion rate, and the higher the priority set accordingly. The formula is as follows: , In the formula, Indicates The task completion degree of each intelligent handling AGV; Indicates The distance travelled by a smart handling AGV.

[0038] The load status indicates the weight of the cargo carried by the intelligent handling AGV when performing a task. It is the ratio of the weight of the cargo carried to the total distance. The larger the ratio, the higher the task completion degree, and the higher the priority set accordingly. The formula is as follows: , In the formula, Indicates The load status of the intelligent handling AGV; Indicates The weight of cargo carried by an intelligent handling AGV.

[0039] The priority of the intelligent handling AGV is composed of these three indicators. The larger the value, the higher the priority of the intelligent handling AGV. The formula is as follows: , in, For the The priority of an intelligent handling AGV.

[0040] In particular, when the energy consumption tasks, task completion degrees, and load states of the conflicting intelligent transport AGVs are equal, the priorities of the intelligent transport AGVs are determined to be the same, and the avoidance party and the priority party are randomly selected.

[0041] In some embodiments, based on the priority of the smart transport AGV that has the time-space conflict and the conflict type of the time-space conflict, a conflict resolution strategy corresponding to the time-space conflict is determined, including: When the conflict type of the time-space conflict is a node conflict, the waiting intelligent transport AGV is determined based on the priority of the intelligent transport AGV with the time-space conflict, and the waiting intelligent transport AGV is temporarily waiting until the node conflict ends. Specifically, the low-priority intelligent transport AGV is suspended and waits until the motion path of the high-priority intelligent transport AGV no longer affects its movement, and then continues to travel along the original path; When the conflict type of the space-time conflict is an opposite conflict, the avoidance intelligent handling AGV is determined based on the priority of the intelligent handling AGV that has the space-time conflict, the avoidance intelligent handling AGV is determined, the avoidance strategy of the avoidance intelligent handling AGV is determined, and the avoidance intelligent handling AGV is scheduled to execute the avoidance strategy. Specifically, for the opposite conflict, the intelligent handling AGVs will definitely collide with each other, and the low-priority pause waiting strategy is no longer applicable. In this case, by judging the priority of each intelligent handling AGV, the intelligent handling AGV with a high priority continues to execute the original path, and the intelligent handling AGV with a low priority performs the avoidance strategy; When the conflict type of the space-time conflict is a same-direction conflict, the waiting smart transport AGV is determined based on the priority of the smart transport AGV where the space-time conflict occurs. The waiting smart transport AGV temporarily waits until the same-direction conflict ends. Specifically, the low-priority smart transport AGV pauses and waits until the motion path of the high-priority smart transport AGV no longer affects its movement, and then continues to travel along the original path.

[0042] Through the above coordination strategies, the movement of intelligent handling AGVs can be effectively coordinated to avoid collisions and conflicts, ensuring the safe and efficient operation of multiple intelligent handling AGVs.

[0043] Step 160 : scheduling multiple intelligent transport AGVs based on the global optimal path of each intelligent transport AGV and the conflict resolution strategy corresponding to each spatiotemporal conflict.

[0044] Specifically include: If the intelligent handling AGV encounters an unknown dynamic obstacle during driving, the speed of the intelligent handling AGV will be adaptively adjusted to perform dynamic obstacle avoidance. If the intelligent transport AGV encounters a time-space conflict during driving, multiple intelligent transport AGVs are dispatched based on the global optimal path of each intelligent transport AGV and the conflict resolution strategy corresponding to the time-space conflict.

[0045] In some embodiments, the speed of the intelligent transport AGV is adaptively adjusted according to the complexity of the environment, including: Based on the improved dynamic window method, the speed of the intelligent handling AGV is adaptively adjusted, specifically including: sampling to obtain a speed combination; simulating multiple motion trajectories based on the sampled speed combination; evaluating each motion trajectory through the evaluation function of the improved dynamic window method to determine the optimal motion trajectory; adjusting the speed of the intelligent handling AGV based on the optimal motion trajectory. For example, the motion model of the intelligent handling AGV is established, taking into account the physical limitations (e.g., maximum and minimum speeds, acceleration, etc.) and safety (e.g., braking distance, etc.) of the intelligent handling AGV, and in the speed space A dynamically changing window is formed in the dynamic window. The speed combination in this window is the speed that the robot can reach in the current state. According to the sampled speed combination, the motion trajectory in the future is predicted through the intelligent handling AGV motion model. The predicted motion trajectory is scored through the evaluation function of the improved dynamic window method, and the motion trajectory with the highest evaluation score is selected as the optimal motion trajectory. The speed of the intelligent handling AGV is adjusted based on the speed corresponding to the optimal motion trajectory.

[0046] In some embodiments, the evaluation function of the improved dynamic window method is: , , , in, is the evaluation score of the motion trajectory, , , and are weight coefficients, is the smoothing coefficient, is the angular deviation between the end point direction of the simulated motion trajectory at the current speed and the target node, is the current estimated running speed, is the path curvature parameter, is the minimum straight-line distance between the end point of the simulated motion trajectory and the known obstacle, is the minimum straight-line distance between the end point of the simulated motion trajectory and the unknown obstacle, For the current critical turning point Axis coordinates, For the current critical turning point Axis coordinates, For the target node Axis coordinates, For the target node Axis coordinates, is the pose angle of the line connecting the current key turning point and the target node in the global coordinate system, The current sub-target point Axis coordinates, For the next sub-target point Axis coordinates, The current sub-target point Axis coordinates, For the next sub-target point Axis coordinates, where the current sub-target point is the key turning point closest to the current key turning point in the forward direction of the current key turning point, and the next sub-target point is the key turning point closest to the current sub-target point in the forward direction of the current sub-target point.

[0047] The following simulation experiments are combined to verify the solving performance of a multi-intelligent handling AGV conflict coordination and dynamic obstacle avoidance collaborative path planning method for maps with different densities. The verification indicators are path length, running time, number of search nodes, number of inflection points and cumulative turning angles.

[0048] Under different obstacle rates and different map sizes, the existing Algorithm, Dijkstra algorithm, improvement The algorithm and the global optimal path planning algorithm recorded in step 130 of this specification (hereinafter referred to as "this method") can both perform global path planning. As shown in Table 1, the performance parameters of four different algorithms under three environments are given, and the environmental obstacle rate , 0.208, 0.082, and 0.245, respectively. These typical maps show different density levels and terrain features. For ease of expression, path length, running time, number of search nodes, number of turning points, and cumulative turning angles are represented by , , , , express.

[0049] Table 1

[0050] In terms of path length, this method reduces the average of 13.03%, 6.79%, and 4.29% compared with the existing A* algorithm, Dijkstra algorithm, and improved A* algorithm, respectively. In terms of running time, this method reduces the average of 74.37%, 71.78%, and 37.05% compared with the existing A* algorithm, Dijkstra algorithm, and improved A* algorithm, respectively. In terms of the number of search nodes, this method reduces the average of 56.45%, 70.67%, and 27.92% compared with the existing A* algorithm, Dijkstra algorithm, and improved A* algorithm, respectively. In terms of the number of inflection points, this method reduces the average of 3.65%, 16.45%, and 18.65% compared with the existing A* algorithm, Dijkstra algorithm, and improved A* algorithm, respectively. In terms of cumulative turning angles, this method reduces the average of 5.45%, 23.67%, and 23.60% compared with the existing A* algorithm, Dijkstra algorithm, and improved A* algorithm, respectively. In addition, with the increase of environmental obstacle rate, the advantages of this method are more obvious.

[0051] Finally, it should be understood that the embodiments described in this specification are only used to illustrate the principles of the embodiments of this specification. Other variations may also fall within the scope of this specification. Therefore, as an example and not a limitation, alternative configurations of the embodiments of this specification may be considered consistent with the teachings of this specification. Accordingly, the embodiments of this specification are not limited to the embodiments explicitly introduced and described in this specification.

Claims

1. A method for collaborative path planning of multi-intelligent handling AGV conflict coordination and dynamic obstacle avoidance. It is characterized in that include: Obtain environmental information, establish a grid map, and set the starting and target nodes of multiple intelligent handling AGVs; Determine the global optimal path of each of the intelligent handling AGVs; Based on the global optimal path of each of the intelligent handling AGVs, determining whether a conflict occurs; If it is determined that a conflict occurs, a multi-AGV path conflict tree is generated based on the global optimal path of each of the intelligent transport AGVs, wherein the multi-AGV path conflict tree is used to record the time-space conflicts on the global optimal paths of multiple intelligent transport AGVs and the conflict types of the time-space conflicts; For each of the space-time conflicts, determine the priority of the intelligent transport AGV that has the space-time conflict, and determine the conflict resolution strategy corresponding to the space-time conflict based on the priority of the intelligent transport AGV that has the space-time conflict and the conflict type of the space-time conflict; The multiple intelligent transport AGVs are scheduled based on the global optimal path of each of the intelligent transport AGVs and the conflict resolution strategy corresponding to each of the spatiotemporal conflicts.

2. According to claim 1, a multi-intelligent transport AGV conflict coordination and dynamic obstacle avoidance collaborative path planning method, It is characterized in that Determine the global optimal path of each of the intelligent handling AGVs, including: S11, establishing an open list and a closed list, and adding the starting point node of the intelligent transport AGV to the open list, wherein the open list is used to record all nodes that have been generated but not fully explored, and the closed list is used to record all nodes that have been explored; S12, introducing the probability of environmental obstacles, adjusting the weight of the heuristic function according to the probability of environmental obstacles, and establishing an adaptive evaluation function; S13, based on the adaptive evaluation function, calculating the evaluation function value of each node in the open list, and determining a visit node from the open list, wherein the visit node is a node in the open list with the smallest evaluation function value; S14. For each adjacent node of the access node, if the adjacent node is not recorded in the closed list and is not recorded in the open list, the adjacent node is added to the open list; if the adjacent node is recorded in the closed list, the adjacent node is ignored; if the adjacent node is not recorded in the closed list and is recorded in the open list, the path length between the access node and the adjacent node is calculated; if the path length between the access node and the adjacent node is shorter than the path length between the parent node of the adjacent node and the adjacent node, the parent node of the adjacent node is updated to be the access node; S15, determining whether the access node is the target node of the intelligent transport AGV or whether the open list is empty, if not, executing S13, if yes, generating the current global path; S16. Introduce a redundant point deletion strategy to perform secondary optimization on the current global path to generate a global optimal path for the intelligent transport AGV.

3. According to claim 2, a multi-intelligent transport AGV conflict coordination and dynamic obstacle avoidance collaborative path planning method, It is characterized in that The adaptive evaluation function is: ， in, For the comprehensive cost, is the actual cost from the starting node to the current node, is the distance between the current node and the target node, is the distance from the starting node to the target node, is the probability of environmental obstacles, is the estimated cost from the current node to the target node.

4. According to claim 2, a multi-intelligent transport AGV conflict coordination and dynamic obstacle avoidance collaborative path planning method, It is characterized in that The introduction of the redundant point deletion strategy performs secondary optimization on the current global path to generate the global optimal path of the intelligent transport AGV, including: S21, taking the next node of the starting point node in the current global path as the current node; S22, determine whether the current node is on the same straight line as the previous node and the parent node of the previous node and has not passed through any obstacles; if the current node is on the same straight line as the previous node and the parent node of the previous node and has not passed through any obstacles, take other nodes on the straight line as redundant nodes and delete the redundant nodes from the current global path; if the current node is not on the same straight line as the previous node and the parent node of the previous node or there is an obstacle on the straight line, take the previous node as a key turning point; S23, determining whether all nodes of the current global path are traversed, if not, executing S24, if yes, generating the global optimal path of the intelligent transport AGV; S24. Take the next node of the current node as the current node and execute S22.

5. According to claim 1, a multi-intelligent handling AGV conflict coordination and dynamic obstacle avoidance collaborative path planning method, It is characterized in that Determine the priority of the intelligent handling AGV that has the time-space conflict, including: The priority of the intelligent transport AGV having the time-space conflict is determined based on the task energy consumption, task completion degree and load status of the intelligent transport AGV having the time-space conflict.

6. A method for collaborative path planning for multi-intelligent transport AGV conflict coordination and dynamic obstacle avoidance according to any one of claims 1 to 5, It is characterized in that Based on the priority of the intelligent transport AGV in which the space-time conflict occurs and the conflict type of the space-time conflict, a conflict resolution strategy corresponding to the space-time conflict is determined, including: When the conflict type of the space-time conflict is a node conflict, the waiting intelligent transport AGV is determined based on the priority of the intelligent transport AGV in which the space-time conflict occurs, and the waiting intelligent transport AGV temporarily waits until the node conflict ends, wherein the node conflict is that at least two of the intelligent transport AGVs will occupy the same node together at the next moment and travel in different directions; When the conflict type of the space-time conflict is an opposite conflict, based on the priority of the intelligent transport AGVs that cause the space-time conflict, determine the intelligent transport AGV to be avoided, determine the avoidance strategy of the intelligent transport AGV to be avoided, and schedule the intelligent transport AGV to execute the avoidance strategy, wherein the opposite conflict is that at least two of the intelligent transport AGVs will jointly occupy each other's path at the next moment and travel in the opposite direction; When the conflict type of the space-time conflict is a same-direction conflict, the waiting smart transport AGV is determined based on the priority of the smart transport AGV where the space-time conflict occurs, and the waiting smart transport AGV temporarily waits until the same-direction conflict ends, wherein the same-direction conflict is that at least two of the smart transport AGVs travel in the same direction after the conflict occurs.

7. A method for collaborative path planning for multi-intelligent transport AGV conflict coordination and dynamic obstacle avoidance according to any one of claims 1 to 5, It is characterized in that Scheduling the multiple intelligent transport AGVs based on the global optimal path of each of the intelligent transport AGVs and the conflict resolution strategy corresponding to each of the spatiotemporal conflicts includes: If the intelligent transport AGV encounters an unknown dynamic obstacle during driving, the speed of the intelligent transport AGV is adaptively adjusted to perform dynamic obstacle avoidance; If the intelligent transport AGV encounters a time-space conflict during driving, the multiple intelligent transport AGVs are scheduled based on the global optimal path of each of the intelligent transport AGVs and the conflict resolution strategy corresponding to the time-space conflict.

8. A method for collaborative path planning for multi-intelligent transport AGV conflict coordination and dynamic obstacle avoidance according to claim 7, It is characterized in that The method of adaptively adjusting the speed of the intelligent transport AGV according to the complexity of the environment includes: The speed of the intelligent transport AGV is adaptively adjusted based on an improved dynamic window method.

9. A method for collaborative path planning for multi-intelligent transport AGV conflict coordination and dynamic obstacle avoidance according to claim 8, It is characterized in that Adaptively adjusting the speed of the intelligent transport AGV based on an improved dynamic window method includes: Sampling to obtain speed combination; Based on the sampled velocity combinations, multiple motion trajectories are simulated; By using the evaluation function of the improved dynamic window method, each of the motion trajectories is evaluated to determine the optimal motion trajectory; Based on the optimal motion trajectory, the speed of the intelligent transport AGV is adjusted.

10. A method for collaborative path planning for multi-intelligent transport AGV conflict coordination and dynamic obstacle avoidance according to claim 9, It is characterized in that The evaluation function of the improved dynamic window method is: ， ， ， in, is the evaluation score of the motion trajectory, , , , are weight coefficients, is the smoothing coefficient, is the angular deviation between the end point direction of the simulated motion trajectory at the current speed and the target node, is the current estimated running speed, Path curvature parameter, is the minimum straight-line distance between the end point of the simulated motion trajectory and the known obstacle, is the minimum straight-line distance between the end point of the simulated motion trajectory and the unknown obstacle, For the current critical turning point Axis coordinates, For the current critical turning point Axis coordinates, For the target node Axis coordinates, For the target node Axis coordinates, is the pose angle of the line connecting the current key turning point and the target node in the global coordinate system, The current sub-target point Axis coordinates, For the next sub-target point Axis coordinates, The current sub-target point Axis coordinates, The next sub-target point Axis coordinates.

Images