Task scheduling method and device based on RGV path planning

By combining the A-star algorithm and CBS conflict resolution method, considering the time-consuming and dynamic obstacles of RGV step size, dynamic update of paths, the problem of inaccurate RGV path planning in the existing technology is solved, and efficient and safe RGV path planning and Scheduling.

CN118963295APending Publication Date: 2024-11-15HUBEI PROLOG TECH CO LTD

Patent Information

Application Number
CN202411149082.6
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Priority Date
2024-05-29
Filing Date
2024-08-21
Publication Date
2024-11-15

AI Technical Summary

Technical Problem

The existing RGV path planning method fails to effectively consider the time-consuming step size of RGV during the movement process and the impact of dynamic obstacles in multi-RGV coordinated scheduling, resulting in inaccurate and unfeasible path planning and the inability to handle scheduling tasks in a timely manner.

Method used

Adopt path planning strategy based on A-star algorithm and CBS conflict resolution method, combine the time-consuming step size of RGV, dynamically update the obstacle set and recalculate the path to ensure that the path does not conflict with the obstacle within the preset time, and improve path planning accuracy and feasibility.

Benefits of technology

By adjusting the path planning in real time, avoiding RGV collisions and obstacle collisions, ensuring the safe and efficient operation of RGV in the working environment, improving task scheduling efficiency and path execution efficiency.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118963295A_ABST
    Figure CN118963295A_ABST
Patent Text Reader

Abstract

The invention provides a task scheduling method and device based on RGV path planning, and relates to the field of warehouse scheduling, and the method comprises the steps: obtaining a current planned path according to an A star algorithm and a current obstacle set under the condition that there is a to-be-executed RGV, and carrying out the task scheduling according to the A star algorithm and the current obstacle set under the condition that the current planned path conflicts with any obstacle; recalculating to obtain an updated current planned path; under the condition that a plurality of RGVs to be executed exist, obtaining a current planned path according to an A star algorithm, a current obstacle set and a CBS conflict solution method, and under the condition that any obstacle in the current planned path conflicts, recalculating to obtain an updated current planned path; and for any RGV to be executed, issuing the updated current plan path corresponding to the RGV to be executed to the RGV to be executed. According to the method, the accuracy and feasibility of path planning are improved, RGV collision and conflict between other obstacles in the path are avoided, and it is ensured that the RGV safely and efficiently operates in the working environment.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of warehouse scheduling, and in particular to a task scheduling method and device based on RGV path planning. Background Art

[0002] In the path planning of Rail Guided Vehicles (RGV), the A-star algorithm commonly used in multi-vehicle scheduling is calculated based on the premise of uniform motion. However, the movement of RGV in actual work is subject to acceleration and deceleration, which cannot lead to a complete match of scheduling. Common path planning is based on a static environment to completely calculate the path nodes to the destination and then send them to the RGV. When the RGV collides with an obstacle during movement, it cannot continue to execute subsequent commands, nor can it trigger a new path planning, or it can directly wait in place. However, its correction process is too long and the scheduling task cannot be processed in time.

[0003] Chinese invention patent CN 117109607 A discloses a multi-AGV path planning method based on path label identification and conflict search, which includes: the flexible manufacturing system central controller plans production tasks according to user orders, generates navigation start and end points for each AGV, and if the task exists in the path trajectory label database, the path is parsed to obtain the path, otherwise a multi-AGV path planning solver based on conflict search (CBS) is used to plan a specific path for each AGV, each AGV senses the local production workshop environment information, performs actions according to the planned path, uses the attribute information of each AGV when performing the path navigation task in the actual workshop, and labels the path trajectory information with data, and updates the path trajectory label database. However, it does not consider the step length time of different RGVs during the movement process, nor does it consider the impact of dynamic obstacles on path planning during the collaborative scheduling of multiple RGVs.

[0004] However, there is currently no technical solution that can take into account the step lengths of different RGVs during movement and efficiently plan path tasks for collaborative operations of multiple RGVs, nor is there a task scheduling method and device based on RGV path planning. Summary of the invention

[0005] The present invention provides a task scheduling method and device based on RGV path planning, which improves the accuracy and feasibility of path planning by combining the step lengths of different RGVs during movement and adopting different path planning strategies for situations where there are different numbers of RGVs to be executed within a preset time.

[0006] In a first aspect, the present invention provides a task scheduling method based on RGV path planning, comprising:

[0007] In the case that there is a pending RGV within the preset duration, the current planned path corresponding to the pending RGV is calculated according to the A-star algorithm and the current obstacle set. In the case that the current planned path conflicts with any obstacle within the first target duration starting from the current moment, the conflict point is marked as a fixed obstacle, the current obstacle set is updated, and the updated current planned path is recalculated until the updated current planned path does not conflict with any obstacle. The first target duration is the time taken for the first two steps of movement corresponding to the pending RGV;

[0008] In the case where there are multiple RGVs to be executed within the preset duration, the current planned path corresponding to any RGV to be executed is calculated according to the A-star algorithm, the current obstacle set and the CBS conflict resolution method. For any current planned path, if there is a conflict with any obstacle in the current planned path within the second target duration starting from the current moment, the conflict point is marked as a fixed obstacle, the current obstacle set is updated, and the updated current planned path is recalculated until there is no conflict with any obstacle in the updated current planned path. The second target duration is the time taken for the first ten steps of movement corresponding to any RGV to be executed;

[0009] For any RGV to be executed, send the updated current planned path corresponding to the RGV to be executed to the RGV to be executed;

[0010] The current obstacle set includes fixed obstacles and mobile obstacles. The fixed obstacles are the points where the RGVs that are currently in an idle state are located, and the mobile obstacles are all path points corresponding to the RGVs that are currently executing tasks.

[0011] According to the task scheduling method based on RGV path planning provided by the present invention, before calculating the current planned path corresponding to the RGV to be executed according to the A-star algorithm and the current obstacle set, the method further includes:

[0012] Every preset time, scan all idle RGVs and all pending tasks corresponding to the current time;

[0013] For each task to be executed, calculate the moving points from any idle RGV to the starting position of the task to be executed, and determine the space RGV with the least moving points as the RGV to be executed corresponding to the task to be executed;

[0014] Traverse all pending tasks in the order in which they are issued, and determine the pending RGV corresponding to each pending task.

[0015] According to the task scheduling method based on RGV path planning provided by the present invention, after determining the RGV to be executed corresponding to each task to be executed, the method further includes:

[0016] When the number of all tasks to be executed is less than the number of all idle RGVs, obtain the current position of any idle RGV;

[0017] When the current position of any idle RGV is on the planned execution path of any task to be executed, a diffusion search is performed with the current position as the center to determine the end point, and the idle RGV is moved from the current point to the end point, where the end point does not conflict with the planned execution path of any task to be executed.

[0018] According to the task scheduling method based on RGV path planning provided by the present invention, the current planned path corresponding to the RGV to be executed is calculated according to the A-star algorithm and the current obstacle set, including:

[0019] According to the A-star algorithm and the current obstacle set, the ordered point set corresponding to any RGV to be executed when executing the task to be executed is calculated;

[0020] The ordered point set is divided to obtain a current planned path and a remaining planned path, wherein the current planned path is a straight-line driving path corresponding to a current calculation starting point to a current calculation end point, and the remaining planned path is a set of other straight-line driving paths.

[0021] According to the task scheduling method based on RGV path planning provided by the present invention, after sending the updated current planned path corresponding to the RGV to be executed to the RGV to be executed, the method further includes:

[0022] Obtaining the next planned path among the remaining planned paths;

[0023] If the next planned path conflicts with any obstacle within the first target duration starting from the next moment, the conflict point is marked as a fixed obstacle, the next obstacle set is updated, and the updated next planned path is recalculated until the updated next planned path does not conflict with any obstacle.

[0024] According to the task scheduling method based on RGV path planning provided by the present invention, before identifying the conflict point as a fixed obstacle, the method further includes:

[0025] For any RGV, obtain the number of moving points corresponding to the RGV in each historical period;

[0026] Determine all unit step times according to all historical durations and the number of moving points corresponding to each historical duration, average all unit step times to obtain the step time when the RGV moves one step, traverse all RGVs, and obtain the step time when each RGV moves one step.

[0027] According to the task scheduling method based on RGV path planning provided by the present invention, the current planned path corresponding to any RGV to be executed is calculated according to the A-star algorithm, the current obstacle set and the CBS conflict resolution method, including:

[0028] According to the A-star algorithm and the current obstacle set, the initial planned path of any RGV to be executed is calculated to construct an initial path set;

[0029] Determine an initial priority queue according to the initial path set and the preset constraint set;

[0030] Traversing all initial planned paths according to the CBS conflict resolution method, determining the constraint conditions of each conflicting party, and re-determining the constraint paths of the nodes corresponding to the conflicting parties according to the constraint conditions;

[0031] The initial priority queue is updated according to the total cost of the constraint path to obtain an updated priority queue, until there is no conflict in the updated priority queue, and a final planned path corresponding to any RGV to be executed is obtained;

[0032] The final planned path is divided to obtain a current planned path corresponding to any RGV to be executed.

[0033] According to the task scheduling method based on RGV path planning provided by the present invention, the updating of the initial priority queue according to the total cost of the constraint path to obtain the updated priority queue includes:

[0034]

[0035] Among them, Cost(N′) is the total cost of the constraint path, Pi′ is the constraint path, Q′ is the updated priority queue, Q is the initial priority queue, Ni′ and Nj′ are the constraints of the conflicting parties.

[0036] According to the task scheduling method based on RGV path planning provided by the present invention, after sending the updated current planned path corresponding to the RGV to be executed to the RGV to be executed, the method further includes:

[0037] When it is determined that the driver of the RGV to be executed is in a linked state with the RGV to be executed, and the updated current planned path conflicts with the current obstacle set, adjusting the end node of the updated current planned path to a point before the collision, and determining each point in the updated current planned path to be a moving obstacle;

[0038] Clear the fixed obstacle corresponding to the point where the RGV in the idle state at the current moment is located in the current obstacle set, and mark the end node of the updated current planned path as a fixed obstacle;

[0039] After the to-be-executed RGV completes the movement of each point in the updated current planned path, the moving obstacle corresponding to the updated current planned path is deleted.

[0040] In a second aspect, a task scheduling device based on RGV path planning is provided, comprising:

[0041] A first service calculation unit is used for, when there is a to-be-executed RGV within a preset duration, calculating the current planned path corresponding to the to-be-executed RGV according to the A-star algorithm and the current obstacle set, and when there is a conflict with any obstacle in the current planned path within a first target duration starting from the current moment, marking the conflict point as a fixed obstacle, updating the current obstacle set, and recalculating the updated current planned path until there is no conflict with any obstacle in the updated current planned path, and the first target duration is the step length time consumption of the first two steps of movement corresponding to the to-be-executed RGV;

[0042] The second service calculation unit is used to calculate the current planned path corresponding to any RGV to be executed according to the A-star algorithm, the current obstacle set and the CBS conflict resolution method when there are multiple RGVs to be executed within a preset time. For any current planned path, if there is a conflict with any obstacle in the current planned path within a second target time starting from the current moment, mark the conflict point as a fixed obstacle, update the current obstacle set, and recalculate the updated current planned path until there is no conflict with any obstacle in the updated current planned path, and the second target time is the time taken for the first ten steps of movement corresponding to any RGV to be executed;

[0043] A scheduling unit, configured to send, for any RGV to be executed, an updated current planned path corresponding to the RGV to be executed to the RGV to be executed;

[0044] The current obstacle set includes fixed obstacles and mobile obstacles. The fixed obstacles are the points where the RGVs that are currently in an idle state are located, and the mobile obstacles are all path points corresponding to the RGVs that are currently executing tasks.

[0045] The present invention adopts different path planning strategies for different task scheduling scenarios at the current moment, fully combines the influence of the step length time of different RGVs in the moving process on the path planning, redefines the formation and removal conditions of dynamic obstacles and fixed obstacles, adjusts the path planning results in real time according to the dynamic motion state of RGV, and efficiently plans the paths of multiple RGVs working in collaboration and the paths of a single RGV working alone, thereby improving the accuracy and feasibility of path planning, avoiding RGV collisions and conflicts with other obstacles in the path, and ensuring the safe and efficient operation of RGV in the working environment;

[0046] The present invention also sets a preset duration, scans all idle RGVs and all tasks to be executed corresponding to the current moment, and plans the path execution efficiency based on the preset path planning frequency; adds fixed obstacles and mobile obstacles to the A* algorithm, plans the execution path parameters, and increases the accuracy of the A* algorithm calculation results; switches the path timing to the overall scheduling to ensure that the path calculation service can obtain the maximum parameter amount;

[0047] The present invention also determines the conflict between the idle RGV and any task to be executed, thereby reducing the re-planning frequency of the planned execution path of the task to be executed, reducing the amount of path planning calculation, and improving the task scheduling efficiency by moving the idle RGV;

[0048] The present invention also calculates the unit step time for RGVs of different manufacturers, different models or different configurations, and averages all the unit step time to obtain the step time when each RGV of different manufacturers, different models or different configurations moves one step, so that in subsequent calculations, the first target time and the second target time can be determined more accurately according to the own properties of different RGVs, thereby improving the accuracy of judging whether the current planned path conflicts with any obstacle, thereby improving the accuracy and effectiveness of path planning. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0050] Figure 1 It is a flowchart of a task scheduling method based on RGV path planning provided by the present invention;

[0051] Figure 2 It is a structural schematic diagram of a task scheduling device based on RGV path planning provided by the present invention;

[0052] Figure 3 It is a structural schematic diagram of the electronic device provided by the present invention. DETAILED DESCRIPTION

[0053] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0054] The existing scheduling system cannot adjust and optimize the dynamic motion state of RGV in real time, resulting in a deviation between the path planning result and the actual operation, which may cause safety hazards such as collision. In addition, due to the acceleration and deceleration of RGV movement, it cannot be terminated in advance and the positioning is inaccurate, resulting in the inability to effectively execute the A-star algorithm. In order to solve the above technical problems, the present invention provides a task scheduling method based on RGV path planning. Figure 1 : is a flow chart of a task scheduling method based on RGV path planning provided by the present invention, the method comprising:

[0055] Step 101: When there is an RGV to be executed within a preset duration, the current planned path corresponding to the RGV to be executed is calculated according to the A-star algorithm and the current obstacle set. When the current planned path conflicts with any obstacle within the first target duration starting from the current moment, the conflict point is marked as a fixed obstacle, the current obstacle set is updated, and the updated current planned path is recalculated until the updated current planned path does not conflict with any obstacle. The first target duration is the time taken for the first two steps of movement corresponding to the RGV to be executed.

[0056] Step 102: When there are multiple RGVs to be executed within the preset duration, the current planned path corresponding to any RGV to be executed is calculated according to the A-star algorithm, the current obstacle set and the CBS conflict resolution method. For any current planned path, when the current planned path conflicts with any obstacle within the second target duration starting from the current moment, the conflict point is marked as a fixed obstacle, the current obstacle set is updated, and the updated current planned path is recalculated until the updated current planned path does not conflict with any obstacle. The second target duration is the time taken for the first ten steps of movement corresponding to any RGV to be executed.

[0057] Step 103: for any RGV to be executed, send the updated current planned path corresponding to the RGV to be executed to the RGV to be executed;

[0058] The current obstacle set includes fixed obstacles and mobile obstacles. The fixed obstacles are the points where the RGVs that are currently in an idle state are located, and the mobile obstacles are all path points corresponding to the RGVs that are currently executing tasks.

[0059] Optionally, before calculating the current planned path corresponding to the to-be-executed RGV according to the A-star algorithm and the current obstacle set, the method further includes:

[0060] Every preset time, scan all idle RGVs and all pending tasks corresponding to the current time;

[0061] For each task to be executed, calculate the moving points from any idle RGV to the starting position of the task to be executed, and determine the space RGV with the least moving points as the RGV to be executed corresponding to the task to be executed;

[0062] Traverse all pending tasks in the order in which they are issued, and determine the pending RGV corresponding to each pending task.

[0063] Specifically, the present invention first initializes the task scheduling system, loads all RGV information and map information into the memory, and stores the current position of the RGV as a fixed obstacle. Then, the present invention can execute the execution process of the task scheduling timer every 1 second, scan all idle RGVs and all tasks to be executed corresponding to the current moment, and then assign all tasks to be executed to the idle RGVs one by one. When obtaining the task list, the present invention can sort the tasks according to the time when they are issued, and give priority to assigning tasks that are issued earlier, so that each task to be executed has a corresponding RGV to be executed.

[0064] Optionally, after determining the RGV to be executed corresponding to each task to be executed, the method further includes:

[0065] When the number of all tasks to be executed is less than the number of all idle RGVs, obtain the current position of any idle RGV;

[0066] When the current position of any idle RGV is on the planned execution path of any task to be executed, a diffusion search is performed with the current position as the center to determine the end point, and the idle RGV is moved from the current point to the end point, where the end point does not conflict with the planned execution path of any task to be executed.

[0067] Optionally, when the number of all tasks to be executed is greater than or equal to the number of all idle RGVs, a task to be executed is first assigned to each idle RGV according to the principle of sequential allocation, and the remaining unassigned tasks to be executed are stored. After any RGV to be executed completes the designated assigned task, tasks from the remaining unassigned tasks to be executed can continue to be assigned to it.

[0068] Optionally, when the number of tasks to be executed is less than the number of all idle RGVs, it is considered that there are still remaining RGVs in an idle state. At this time, the remaining RGVs are stored as fixed obstacles, and the current position of any idle RGV is obtained. When the current position of any idle RGV is on the planned execution path of any task to be executed, in order to ensure that the RGV currently executing the task can more conveniently execute the scheduling task, the present invention schedules and moves the idle RGV, and checks whether the idle RGV is on the planned execution path of other RGVs. If so, an avoidance task is issued to the idle RGV for the conflicting vehicle. The calculation logic of the end position of the avoidance task is: starting from the current position of the RGV, diffusely search for a point that does not conflict with the planned paths of other RGVs as the end position, that is, the present invention diffuses and searches with the current point as the center, determines the end point, and moves the idle RGV from the current point to the end point, and the end point is a point that does not conflict with the planned execution path of any task to be executed.

[0069] In step 101, the present invention adopts a path planning strategy combining the A-star algorithm and the current obstacle set to obtain the current planned path corresponding to the RGV to be executed in a preset period.

[0070] Optionally, the step of calculating the current planned path corresponding to the RGV to be executed according to the A-star algorithm and the current obstacle set includes:

[0071] According to the A-star algorithm and the current obstacle set, the ordered point set corresponding to any RGV to be executed when executing the task to be executed is calculated;

[0072] The ordered point set is divided to obtain a current planned path and a remaining planned path, wherein the current planned path is a straight-line driving path corresponding to a current calculation starting point to a current calculation end point, and the remaining planned path is a set of other straight-line driving paths.

[0073] Optionally, the algorithm for path planning in a dynamic environment is particularly suitable for problems where the environmental state changes over time, such as RGV path planning, vehicle scheduling in a traffic system, etc. The traditional A algorithm mainly focuses on the shortest path problem in a static environment, while the present invention expands this point by taking the time dimension into account. The state not only contains the spatial position but also contains the time information, that is, the state can be expressed as (x, y, t), where (x, y) is the spatial coordinate and t is the time. The cost function represents the cumulative cost from the starting point to the current state. In the present invention, the cost includes not only the cost of spatial movement but also the cost in time. A priority queue is used to store and select the expansion node, and the priority is usually based on the estimated total cost. By considering the time dimension, the present invention inserts the starting state into the priority queue, sets its cost and the estimated total cost, and takes out the node with the smallest estimated total cost from the priority queue for expansion. For each possible successor node of the current node, its cost is calculated and the priority queue is updated. If the target state is reached, the path is terminated and returned, and the final path is constructed by backtracking from the target state to the starting state.

[0074] After the present invention completes the calculation according to the A-star algorithm and the current obstacle set, the obtained path planning result is an ordered point set, and the point set is cut according to whether the X and Y axes are consistent and whether the points are retreated, for example (0, 1), (0, 2), (0, 3), (1, 3), (2, 3), (1, 3), (1, 4), (1, 5), and the above ordered points are converted into the following commands (0, 1)-(0, 3), (1, 3)-(2, 3), (2, 3)-(1, 3), (1, 3)-(1, 5), wherein the first section of commands (0, 1)-(0, 3) is the current planned path, and other commands and the points contained therein are stored as the remaining planned paths.

[0075] Optionally, the present invention determines the step time according to the time required for the first two steps of movement corresponding to the RGV to be executed, and uses the step time as the first target time, and judges whether the RGV to be executed will conflict with any obstacle when executing the current planned path within the first target time. If no conflict occurs, the subsequent planned path continues to be executed. If a conflict occurs, it is considered that the RGV to be executed encounters an obstacle when executing the current planned path, and the conflict point is marked as a fixed obstacle. The fixed obstacle is added to the current obstacle set to update the current obstacle set. Since the current obstacle set has changed, the updated ordered point set corresponding to any RGV to be executed when executing the task to be executed is calculated according to the A-star algorithm and the updated current obstacle set, and then the updated ordered point set is re-divided to obtain the updated current planned path, and whether the updated current planned path will conflict with any obstacle is judged again until there is no conflict with any obstacle in the updated current planned path. The first target time is the step time of the first two steps of movement corresponding to the RGV to be executed.

[0076] Optionally, after sending the updated current planned path corresponding to the to-be-executed RGV to the to-be-executed RGV, the method further includes:

[0077] Obtaining the next planned path among the remaining planned paths;

[0078] If the next planned path conflicts with any obstacle within the first target duration starting from the next moment, the conflict point is marked as a fixed obstacle, the next obstacle set is updated, and the updated next planned path is recalculated until the updated next planned path does not conflict with any obstacle.

[0079] Optionally, after executing the updated current planned path, RGV only moves the first segment of the planned path, and for each planned path in the remaining planned paths, it can be determined whether each planned path needs to be recalculated and whether the planned path can be updated based on the above-mentioned conflict judgment mechanism. For example, if there is a conflict with any obstacle in the next planned path, the conflict point is marked as a fixed obstacle, the next obstacle set is updated, and the updated ordered point set is calculated again based on the A-star algorithm and the updated current obstacle set to obtain the updated next planned path, until there is no conflict with any obstacle in the updated next planned path.

[0080] Optionally, before identifying the conflict point as a fixed obstacle, the method further includes:

[0081] For any RGV, obtain the number of moving points corresponding to the RGV in each historical period;

[0082] Determine all unit step times according to all historical durations and the number of moving points corresponding to each historical duration, average all unit step times to obtain the step time when the RGV moves one step, traverse all RGVs, and obtain the step time when each RGV moves one step.

[0083] Since each RGV has different factory settings and models, the step time of RGVs with different configurations and models when moving one step is also different. The present invention can pre-acquire the time spent and the distance moved by each RGV in the historical movement process, average the unit step time corresponding to each historical movement, and then obtain the step time required for the RGV to move one step, and then traverse all RGVs to obtain the step time of each RGV when moving one step, so that in subsequent calculations, the first target time and the second target time can be determined more accurately according to the own properties of different RGVs, thereby improving the accuracy of the judgment of whether the current planned path conflicts with any obstacle, thereby improving the accuracy and effectiveness of path planning.

[0084] In step 102, the CBS conflict resolution method is a conflict-based MAPF algorithm. CBS gives the global optimal result of the MAPF problem. Its algorithm idea is to divide the multi-machine planning into two layers. The bottom layer executes the single-machine planning with constraints. For example, the top layer of the traditional A-star algorithm traverses the planning path of the bottom layer to check whether there is a conflict between the paths. If there is a conflict, the constraints are imposed to re-plan the bottom-layer single-machine until all the bottom-layer paths are conflict-free. The present invention adds a moving step on the basis of the traditional A-star algorithm to add obstacles in fixed steps. Since RGV movement involves acceleration and deceleration, the traditional A-star algorithm does not apply the method of avoiding conflicts through moving steps. Therefore, the moving step is first converted into a moving instruction, and then the point is locked for a time period according to the step length. The moving instruction refers to the movement of the RGV according to the coordinates, not the point movement. For example, the moving covering points (0, 1), (0, 2), and (0, 3) are sent to the RGV, and the corresponding instruction is (0, 3).

[0085] Optionally, in the case where there are multiple RGVs to be executed within a preset period, the present invention adopts a path planning strategy combining the A-star algorithm, the current obstacle set and the CBS conflict resolution method to obtain the current planned path corresponding to any RGV to be executed. The current planned path corresponding to any RGV to be executed is calculated based on the A-star algorithm, the current obstacle set and the CBS conflict resolution method, including:

[0086] Calculate the initial planned path of any RGV to be executed according to the A-star algorithm and the current obstacle set, and construct an initial path set;

[0087] Determine an initial priority queue according to the initial path set and the preset constraint set;

[0088] Traversing all initial planned paths according to the CBS conflict resolution method, determining the constraint conditions of each conflicting party, and re-determining the constraint paths of the nodes corresponding to the conflicting parties according to the constraint conditions;

[0089] The initial priority queue is updated according to the total cost of the constraint path to obtain an updated priority queue, until there is no conflict in the updated priority queue, and a final planned path corresponding to any RGV to be executed is obtained;

[0090] The final planned path is divided to obtain a current planned path corresponding to any RGV to be executed.

[0091] Optionally, a formula can be used to describe the calculation process of the CBS (Conflict-Based Search) algorithm, especially to describe how to handle paths and conflicts in high-level and low-level searches. In path initialization, the path Pi of each RGVi is calculated by an independent path planning algorithm, such as the A-star algorithm. The initial path set is: P = {P1, P2, ..., Pn}, where n is the number of RGVs.

[0092] Optionally, in a high-level search, each node N contains an initial path set P and a set of preset constraints C, and the initial node Nroot is represented as: Nroot = (P, C) where The initial priority queue Q is sorted according to the path cost and initially contains the root node: Q = {Nroot}.

[0093] Optionally, iterate over all paths, looking for conflicts: Where Pi(t) represents the position of RGVi at time t. If a conflict is found, record the conflict tuple: Conflict(Pi, Pj, t) = (i, j, Pi(t), t). When handling conflicts, for each conflict, two new nodes are generated and corresponding constraints are added to each node. For RGVi, add constraint (i, Pi(t), t); for RGVj, add constraint (j, Pj(t), t).

[0094] Optionally, the new node N′ contains updated paths and constraints, Ni′=(P′, C∪{(i, Pi(t), t)}); Nj′=(P′, C∪{(j, Pj(t), t)}). In the underlying search, the path is recalculated, and the path is recalculated for the affected RGV under the new constraints to redetermine the constraint paths of the corresponding nodes of the conflicting parties according to the constraints, and the A-star algorithm is used to calculate the paths Pi′ and Pj′ that satisfy the new constraints, where Pi′=A*(i, starti, goali, C∪{(i, Pi(t), t)}); Pj′=A*(j, startj, goalj, C∪{(j, Pj(t), t)}).

[0095] Optionally, updating the initial priority queue according to the total cost of the constrained path to obtain an updated priority queue includes:

[0096]

[0097] Among them, Cost(N′) is the total cost of the constraint path, Pi′ is the constraint path, Q′ is the updated priority queue, Q is the initial priority queue, Ni′ and Nj′ are the constraints of the conflicting parties.

[0098] Optionally, repeat the above-mentioned operation of updating the priority queue until there is no conflict in the updated priority queue, obtain the final planned path corresponding to any RGV to be executed, split the final planned path, and obtain the current planned path and remaining planned paths corresponding to any RGV to be executed.

[0099] Optionally, the first target duration is the step length of the first two steps of movement corresponding to the RGV to be executed, and the second target duration is the step length of the first ten steps of movement corresponding to any RGV to be executed. The present invention sets the second target duration as the step length of the first ten steps of movement corresponding to any RGV to be executed in order to better determine the most suitable current planned path in CBS conflict resolution. Since acceleration and deceleration errors will be superimposed in movement, CBS only calculates the first 10 steps when calculating conflicts to avoid over-calculation. By switching the path calculation timing to overall scheduling, the path calculation can adjust the planned route in time.

[0100] In step 103, for any RGV to be executed, the updated current planned path corresponding to the RGV to be executed is sent to the RGV to be executed. After sending the updated current planned path corresponding to the RGV to be executed to the RGV to be executed, the method further includes:

[0101] When it is determined that the driver of the RGV to be executed is in a linked state with the RGV to be executed, and the updated current planned path conflicts with the current obstacle set, adjusting the end node of the updated current planned path to a point before the collision, and determining each point in the updated current planned path to be a moving obstacle;

[0102] Clear the fixed obstacle corresponding to the point where the RGV in the idle state at the current moment is located in the current obstacle set, and mark the end node of the updated current planned path as a fixed obstacle;

[0103] After the to-be-executed RGV completes the movement of each point in the updated current planned path, the moving obstacle corresponding to the updated current planned path is deleted.

[0104] Optionally, when the RGV control receives a moving instruction, it checks whether the RGV driver and the RGV device are linked, and checks again whether there is a conflict between the command to be executed and the fixed obstacles and the moving obstacles. If so, the end position of the command is modified to the point before the collision point, and the newly generated command is converted into a moving obstacle. The cache is locked by area during the inspection process to avoid calculation anomalies caused by concurrency. Further, the command is sent to the RGV device for execution, the fixed obstacles corresponding to the current RGV in the memory are cleared, and the end position of the execution command is stored as a new fixed obstacle. The command contains the point as a moving obstacle. During the execution of the command, the moving obstacles corresponding to the executed points in the memory are cleared, and the moving obstacles corresponding to the RGV are deleted when the command ends. After the command is executed, the list of RGVs to be planned is entered.

[0105] The present invention increases the efficiency of path execution by presetting the path planning frequency; by adding fixed obstacles and mobile obstacles to the A* algorithm, the execution path parameters are planned to increase the accuracy of the calculation results of the A* algorithm; by switching the path timing to the overall scheduling to ensure that the path calculation service can obtain the maximum parameter amount, different path planning strategies are adopted for different task scheduling scenarios at the current moment, and the influence of the step length time consumption of different RGVs in the moving process on the path planning is fully combined, the formation and removal conditions of dynamic obstacles and fixed obstacles are redefined, the path planning results are adjusted in real time according to the dynamic motion state of the RGV, and the paths of multiple RGVs working in collaboration and the paths of a single RGV working alone are efficiently planned, thereby improving the accuracy and feasibility of path planning, avoiding RGV collisions and conflicts with other obstacles in the path, and ensuring the safe and efficient operation of the RGV in the working environment.

[0106] Figure 2It is a structural schematic diagram of a task scheduling device based on RGV path planning provided by the present invention, wherein the task scheduling device based on RGV path planning comprises a first service computing unit 1, which is used for, when there is a to-be-executed RGV within a preset duration, calculating the current planned path corresponding to the to-be-executed RGV according to the A-star algorithm and the current obstacle set, and when there is a conflict with any obstacle in the current planned path within the first target duration starting from the current moment, marking the conflict point as a fixed obstacle, updating the current obstacle set, and recalculating the updated current planned path until there is no conflict with any obstacle in the updated current planned path, and the first target duration is the step length time consumption of the first two steps corresponding to the to-be-executed RGV. The working principle of the first service computing unit 1 can refer to the aforementioned step 101, which will not be repeated here.

[0107] The task scheduling device based on RGV path planning also includes a second service calculation unit 2, which is used to calculate the current planned path corresponding to any RGV to be executed according to the A-star algorithm, the current obstacle set and the CBS conflict resolution method when there are multiple RGVs to be executed within a preset time. For any current planned path, if the current planned path conflicts with any obstacle within the second target time starting from the current moment, the conflict point is marked as a fixed obstacle, the current obstacle set is updated, and the updated current planned path is recalculated until the updated current planned path does not conflict with any obstacle. The second target time is the step length time of the first ten steps corresponding to any RGV to be executed. The working principle of the second service calculation unit 2 can refer to the aforementioned step 102 and will not be repeated here.

[0108] The task scheduling device based on RGV path planning also includes a scheduling unit 3, which is used to send the updated current planned path corresponding to any RGV to be executed to the RGV to be executed. The working principle of the scheduling unit 3 can refer to the aforementioned step 103 and will not be repeated here.

[0109] The current obstacle set includes fixed obstacles and mobile obstacles. The fixed obstacles are the points where the RGVs that are currently in an idle state are located, and the mobile obstacles are all path points corresponding to the RGVs that are currently executing tasks.

[0110] The present invention can obtain the dynamic operation information of RGV in real time by scanning the RGV status at regular intervals. The path planning service module performs dynamic path planning based on the information, which can effectively reduce the deviation between the path planning result and the actual operation and avoid potential safety hazards such as collision. The present invention can maintain the point information, other RGV status and other dynamic obstacle information. The path planning service uses this information for planning, which can fully consider the influence of dynamic obstacles and make the planned path more in line with the actual situation. During the scheduling process, the current path planning and the remaining path planning of the RGV to be executed realize the high efficiency of path planning when multiple RGVs work together, avoiding collisions and path conflicts between RGVs. Tasks can be reasonably allocated to idle RGVs according to actual conditions, thereby improving the utilization efficiency of RGVs. It is conducive to the expansion and maintenance of the system, such as corresponding driver modules can be written for RGVs from different manufacturers.

[0111] The present invention adopts different path planning strategies for different task scheduling scenarios at the current moment, fully combines the impact of the step length time of different RGVs in the moving process on path planning, redefines the formation and removal conditions of dynamic obstacles and fixed obstacles, adjusts the path planning results in real time according to the dynamic motion state of the RGV, and efficiently plans the paths of multiple RGVs working in collaboration and the path of a single RGV working alone, thereby improving the accuracy and feasibility of path planning, avoiding RGV collisions and conflicts with other obstacles in the path, and ensuring the safe and efficient operation of the RGV in the working environment.

[0112] Figure 3 Schematic diagram of the structure of the electronic device provided by the present invention. Figure 3As shown, the electronic device may include: a processor (processor) 310, a communication interface (Communications Interface) 320, a memory (memory) 330 and a communication bus 340, wherein the processor 310, the communication interface 320, and the memory 330 communicate with each other through the communication bus 340. The processor 310 can call the logic instructions in the memory 330 to execute the task scheduling method based on RGV path planning, which includes: when there is a to-be-executed RGV within a preset time length, the current planned path corresponding to the to-be-executed RGV is calculated according to the A star algorithm and the current obstacle set; when the current planned path conflicts with any obstacle within the first target time length starting from the current moment, the conflict point is marked as a fixed obstacle, the current obstacle set is updated, and the updated current planned path is recalculated until the updated current planned path does not conflict with any obstacle, and the first target time length is the step length time of the first two steps corresponding to the to-be-executed RGV; when there are multiple to-be-executed RGVs within the preset time length, according to the A star algorithm, the current obstacle set and the CBS conflict resolution method, the The current planned path corresponding to any RGV to be executed, for any current planned path, if the current planned path conflicts with any obstacle within the second target time starting from the current moment, the conflict point is marked as a fixed obstacle, the current obstacle set is updated, and the updated current planned path is recalculated until there is no conflict with any obstacle in the updated current planned path, and the second target time is the time taken by the first ten steps of movement corresponding to any RGV to be executed; for any RGV to be executed, the updated current planned path corresponding to the RGV to be executed is sent to the RGV to be executed; the current obstacle set includes fixed obstacles and mobile obstacles, the fixed obstacle is the point where the RGV is in an idle state at the current moment, and the mobile obstacle is all path points corresponding to the RGV that is executing the task at the current moment.

[0113] In addition, the logic instructions in the above-mentioned memory 330 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.

[0114] On the other hand, the present invention also provides a computer program product, which includes a computer program, which can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute a task scheduling method based on RGV path planning provided by the above methods, the method including: when there is an RGV to be executed within a preset time length, the current planned path corresponding to the RGV to be executed is calculated according to the A-star algorithm and the current obstacle set; when the current planned path conflicts with any obstacle within the first target time length starting from the current moment, the conflict point is marked as a fixed obstacle, the current obstacle set is updated, and the updated current planned path is recalculated until the updated current planned path does not conflict with any obstacle, the first target time length is the step length time of the first two steps corresponding to the RGV to be executed; when there are multiple RGVs to be executed within the preset time length In this case, the current planned path corresponding to any RGV to be executed is calculated according to the A-star algorithm, the current obstacle set and the CBS conflict resolution method. For any current planned path, if there is a conflict with any obstacle in the current planned path within the second target time starting from the current moment, the conflict point is marked as a fixed obstacle, the current obstacle set is updated, and the updated current planned path is recalculated until there is no conflict with any obstacle in the updated current planned path, and the second target time is the time taken by the first ten steps of movement corresponding to any RGV to be executed; for any RGV to be executed, the updated current planned path corresponding to the RGV to be executed is sent to the RGV to be executed; the current obstacle set includes fixed obstacles and mobile obstacles, the fixed obstacle is the point where the RGV in the idle state is located at the current moment, and the mobile obstacle is all path points corresponding to the RGV that is executing the task at the current moment.

[0115] On the other hand, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which is implemented when the computer program is executed by the processor to execute the task scheduling method based on RGV path planning provided by the above-mentioned methods, the method comprising: when there is one RGV to be executed within a preset time length, calculating the current planned path corresponding to the RGV to be executed according to the A star algorithm and the current obstacle set, and when there is a conflict with any obstacle in the current planned path within a first target time length starting from the current moment, marking the conflict point as a fixed obstacle, updating the current obstacle set, and recalculating the updated current planned path until there is no conflict with any obstacle in the updated current planned path, and the first target time length is the step length time of the first two steps corresponding to the RGV to be executed; when there are multiple RGVs to be executed within the preset time length, calculating the current planned path corresponding to the RGV to be executed according to the A star algorithm and the current obstacle set The object set and CBS conflict resolution method are used to calculate the current planned path corresponding to any RGV to be executed. For any current planned path, if the current planned path conflicts with any obstacle within the second target time starting from the current moment, the conflict point is marked as a fixed obstacle, the current obstacle set is updated, and the updated current planned path is recalculated until there is no conflict with any obstacle in the updated current planned path. The second target time is the time taken for the first ten steps of movement corresponding to any RGV to be executed. For any RGV to be executed, the updated current planned path corresponding to the RGV to be executed is sent to the RGV to be executed. The current obstacle set includes fixed obstacles and mobile obstacles. The fixed obstacle is the point where the RGV is in an idle state at the current moment, and the mobile obstacle is all path points corresponding to the RGV that is currently executing the task.

[0116] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0117] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0118] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A task scheduling method based on RGV path planning, characterized in that: include: In the case that there is a pending RGV within the preset duration, the current planned path corresponding to the pending RGV is calculated according to the A-star algorithm and the current obstacle set. In the case that the current planned path conflicts with any obstacle within the first target duration starting from the current moment, the conflict point is marked as a fixed obstacle, the current obstacle set is updated, and the updated current planned path is recalculated until the updated current planned path does not conflict with any obstacle. The first target duration is the time taken for the first two steps of movement corresponding to the pending RGV; In the case where there are multiple RGVs to be executed within the preset duration, the current planned path corresponding to any RGV to be executed is calculated according to the A-star algorithm, the current obstacle set and the CBS conflict resolution method. For any current planned path, if there is a conflict with any obstacle in the current planned path within the second target duration starting from the current moment, the conflict point is marked as a fixed obstacle, the current obstacle set is updated, and the updated current planned path is recalculated until there is no conflict with any obstacle in the updated current planned path. The second target duration is the time taken for the first ten steps of movement corresponding to any RGV to be executed; For any RGV to be executed, send the updated current planned path corresponding to the RGV to be executed to the RGV to be executed; The current obstacle set includes fixed obstacles and mobile obstacles. The fixed obstacles are the points where the RGVs that are currently in an idle state are located, and the mobile obstacles are all path points corresponding to the RGVs that are currently executing tasks.

2. The task scheduling method based on RGV path planning according to claim 1 is characterized in that: Before calculating the current planned path corresponding to the RGV to be executed according to the A-star algorithm and the current obstacle set, the method further includes: Every preset time, scan all idle RGVs and all pending tasks corresponding to the current time; For each task to be executed, calculate the moving points from any idle RGV to the starting position of the task to be executed, and determine the space RGV with the least moving points as the RGV to be executed corresponding to the task to be executed; Traverse all pending tasks in the order in which they are issued, and determine the pending RGV corresponding to each pending task.

3. The task scheduling method based on RGV path planning according to claim 2 is characterized in that: After determining the RGV to be executed corresponding to each task to be executed, the method further includes: When the number of all tasks to be executed is less than the number of all idle RGVs, obtain the current position of any idle RGV; When the current position of any idle RGV is on the planned execution path of any task to be executed, a diffusion search is performed with the current position as the center to determine the end point, and the idle RGV is moved from the current point to the end point, where the end point does not conflict with the planned execution path of any task to be executed.

4. The task scheduling method based on RGV path planning according to claim 1 is characterized in that: The current planned path corresponding to the RGV to be executed is calculated according to the A-star algorithm and the current obstacle set, including: According to the A-star algorithm and the current obstacle set, the ordered point set corresponding to any RGV to be executed when executing the task to be executed is calculated; The ordered point set is divided to obtain a current planned path and a remaining planned path, wherein the current planned path is a straight-line driving path corresponding to a current calculation starting point to a current calculation end point, and the remaining planned path is a set of other straight-line driving paths.

5. The task scheduling method based on RGV path planning according to claim 4 is characterized in that: After sending the updated current planned path corresponding to the RGV to be executed to the RGV to be executed, the method further includes: Obtaining the next planned path among the remaining planned paths; If the next planned path conflicts with any obstacle within the first target duration starting from the next moment, the conflict point is marked as a fixed obstacle, the next obstacle set is updated, and the updated next planned path is recalculated until the updated next planned path does not conflict with any obstacle.

6. The task scheduling method based on RGV path planning according to claim 1 is characterized in that: Before identifying the conflict point as a fixed obstacle, the method further includes: For any RGV, obtain the number of moving points corresponding to the RGV in each historical period; Determine all unit step times according to all historical durations and the number of moving points corresponding to each historical duration, average all unit step times to obtain the step time when the RGV moves one step, traverse all RGVs, and obtain the step time when each RGV moves one step.

7. The task scheduling method based on RGV path planning according to claim 1 is characterized in that: The current planned path corresponding to any RGV to be executed is calculated according to the A-star algorithm, the current obstacle set and the CBS conflict resolution method, including: Calculate the initial planned path of any RGV to be executed according to the A-star algorithm and the current obstacle set, and construct an initial path set; Determine an initial priority queue according to the initial path set and the preset constraint set; Traversing all initial planned paths according to the CBS conflict resolution method, determining the constraint conditions of each conflicting party, and re-determining the constraint paths of the nodes corresponding to the conflicting parties according to the constraint conditions; The initial priority queue is updated according to the total cost of the constraint path to obtain an updated priority queue, until there is no conflict in the updated priority queue, and a final planned path corresponding to any RGV to be executed is obtained; The final planned path is divided to obtain a current planned path corresponding to any RGV to be executed.

8. The task scheduling method based on RGV path planning according to claim 7 is characterized in that: The updating of the initial priority queue according to the total cost of the constrained path to obtain an updated priority queue includes: Among them, Cost(N′) is the total cost of the constraint path, Pi′ is the constraint path, Q′ is the updated priority queue, Q is the initial priority queue, Ni′ and Nj′ are the constraints of the conflicting parties.

9. The task scheduling method based on RGV path planning according to claim 1 is characterized in that: After sending the updated current planned path corresponding to the RGV to be executed to the RGV to be executed, the method further includes: When it is determined that the driver of the RGV to be executed is in a linked state with the RGV to be executed, and the updated current planned path conflicts with the current obstacle set, adjusting the end node of the updated current planned path to a point before the collision, and determining each point in the updated current planned path to be a moving obstacle; Clear the fixed obstacle corresponding to the point where the RGV in the idle state at the current moment is located in the current obstacle set, and mark the end node of the updated current planned path as a fixed obstacle; After the to-be-executed RGV completes the movement of each point in the updated current planned path, the moving obstacle corresponding to the updated current planned path is deleted.

10. A task scheduling device based on RGV path planning, characterized in that: include: A first service calculation unit is used for, when there is a to-be-executed RGV within a preset duration, calculating the current planned path corresponding to the to-be-executed RGV according to the A-star algorithm and the current obstacle set, and when there is a conflict with any obstacle in the current planned path within a first target duration starting from the current moment, marking the conflict point as a fixed obstacle, updating the current obstacle set, and recalculating the updated current planned path until there is no conflict with any obstacle in the updated current planned path, and the first target duration is the step length time consumption of the first two steps of movement corresponding to the to-be-executed RGV; The second service calculation unit is used to calculate the current planned path corresponding to any RGV to be executed according to the A-star algorithm, the current obstacle set and the CBS conflict resolution method when there are multiple RGVs to be executed within a preset time. For any current planned path, if there is a conflict with any obstacle in the current planned path within a second target time starting from the current moment, mark the conflict point as a fixed obstacle, update the current obstacle set, and recalculate the updated current planned path until there is no conflict with any obstacle in the updated current planned path, and the second target time is the time taken for the first ten steps of movement corresponding to any RGV to be executed; A scheduling unit, configured to send, for any RGV to be executed, an updated current planned path corresponding to the RGV to be executed to the RGV to be executed; The current obstacle set includes fixed obstacles and mobile obstacles. The fixed obstacles are the points where the RGVs that are currently in an idle state are located, and the mobile obstacles are all path points corresponding to the RGVs that are currently executing tasks.

Citation Information

Patent Citations

  • Multi-AGV path planning method based on path label identification and conflict search

    CN117109607A

Cited By

  • Workshop equipment intelligent scheduling method, equipment and medium

    CN120993845A

  • A method, equipment and medium for intelligent scheduling of workshop equipment

    CN120993845B

  • RGV path planning method and system

    CN121411456A