Automatic guided vehicle collision avoidance method and device, computer equipment and storage medium

Abstract

The invention provides an automatic guided vehicle collision avoidance method and device, computer equipment and a storage medium, and belongs to the field of automatic control, and the method comprises the steps: constructing a grid model of a target region; introducing a direction function into the ant colony algorithm to obtain an improved ant colony algorithm, and planning a grid path of the automated guided vehicle in the target area through the improved ant colony algorithm; time windows corresponding to different grids of each automatic guided vehicle in the grid path are determined; and under the condition that the time windows of different automated guided vehicles are overlapped or conflicted, expanding the time nodes of the improved ant colony algorithm, and re-planning the grid paths of the automated guided vehicles corresponding to the overlapped or conflicted time windows. By introducing the direction function to optimize the ant colony algorithm, the search efficiency and convergence speed of the ant colony algorithm are improved, the planning efficiency of the grid path is improved, and the path of the AGV can be dynamically adjusted in combination with the time window, so that more flexible and efficient collision avoidance planning is realized, and the delay and downtime are reduced.

Description

Technical Field

[0001] The present invention belongs to the field of automatic control, and in particular relates to a collision avoidance method and device for an automatic guided vehicle, a computer device and a storage medium. Background Art

[0002] As manufacturing companies rapidly advance towards building smart factories through digitalization, networking, and intelligentization, the need for discrete manufacturing companies to upgrade and transform their factory floors has become inevitable. As a crucial component of the entire production process, the production logistics system is responsible for multiple tasks, including material transportation, storage, loading and unloading. It serves as the organic link between all aspects of production and processing equipment and is fundamental to maintaining the production process. Therefore, automated guided vehicles (AGVs), with their advantages of precise and efficient transportation, ease of management and scheduling, unmanned transportation, and guaranteed labor safety, have been widely adopted in various manufacturing fields.

[0003] In the real world, large factories handle large orders, requiring multiple AGVs (Automated Guided Vehicles) to operate together to efficiently transport products. In a dynamic multi-AGV system, logistics and transportation are complex, and as orders increase and more AGVs are added, path planning becomes even more complex. Consequently, system failures, improper scheduling, or unexpected situations can lead to AGV collisions, path conflicts, and deadlocks.

[0004] Existing technologies pre-plan paths for all AGVs by comprehensively considering the environment map and AGV status information. This path planning method can achieve a globally optimal solution and is highly resistant to interference. However, as the scale of the environment and the number of AGVs increase, the computational complexity of this method increases dramatically, making it difficult to quickly and efficiently obtain the optimal solution. Summary of the Invention

[0005] In order to solve the above-mentioned problem of difficulty in quickly and efficiently obtaining the optimal solution for the AGV path, the present invention provides an automated guided vehicle collision avoidance method, device, computer equipment and storage medium.

[0006] In order to achieve the above object, the present invention provides the following technical solutions: The present invention provides a collision avoidance method for an automated guided vehicle, the method comprising: Construct a grid model of the target area; A direction function is introduced into the ant colony algorithm to obtain an improved ant colony algorithm, a direction of a path is specified by the improved ant colony algorithm, and a path is planned for the automatic guided vehicle on the grid model based on the specified direction to obtain a grid path; Determining a time window of each automated guided vehicle in the grid path and time nodes corresponding to different grids in the time window; In the case where the time windows of different automated guided vehicles overlap or conflict, the time nodes are adjusted and the automated guided vehicles corresponding to the overlapping or conflicting time windows are replanned.

[0007] Optionally, constructing the grid model of the target area includes: obtaining a plane figure of the target area; determining the linear moving distance of the automatic guided vehicle per unit time as the grid side length, dividing the plane figure into a plurality of grids, and obtaining the grid model.

[0008] Optionally, the direction function is introduced into the ant colony algorithm to obtain an improved ant colony algorithm including: determining the probability rule of the kth ant moving from point i to point j at time t for: ; in is the concentration of pheromone, is the heuristic function, α is the pheromone heuristic factor, β is the expected heuristic factor, represents all possible locations that ant k may pass through when moving; i and j represent two different locations respectively; Introducing a direction function , the specific formula is as follows: ; The improved ant movement probability rule is as follows: ; in, is the angle between the direction from the starting point to the end point and the horizontal axis in the coordinate system constructed based on the grid model, is the angle between the ant from point i to point j and the horizontal axis, is the angle between the ant’s current forward direction and the direction from the starting point to the end point, The values are the angle parameter and the direction parameter respectively.

[0009] Optionally, the time window model is expressed as: ; in, For any AGV, specify the time window corresponding to the path in the grid model. is the time it takes for the AGV to pass through any grid in the specified path, It is the time point when the AGV head enters any of the grids. It is the time point when the tail of the AGV leaves any grid.

[0010] Optionally, adjusting the time nodes and replanning the automated guided vehicles corresponding to the overlapping or conflicting time windows includes: determining the priority of the automated guided vehicles corresponding to the overlapping or conflicting time windows; adjusting the time windows for automated guided vehicles with lower priorities so that they avoid the overlapping or conflicting time windows; or replanning the grids corresponding to the overlapping or conflicting time nodes so that automated guided vehicles with lower priorities detour or wait.

[0011] Optionally, the priority of the automated guided vehicle is determined by the mission urgency, grid path efficiency and current operating status of the automated guided vehicle.

[0012] The present invention also provides a collision avoidance device for an automated guided vehicle, the device comprising: A construction module is used to construct a grid model of the target area; a planning module, configured to introduce a direction function into the ant colony algorithm to obtain an improved ant colony algorithm, specify a path direction using the improved ant colony algorithm, and perform path planning for the automated guided vehicle on the grid model based on the specified direction to obtain a grid path; a determination module, configured to determine a time window of each automated guided vehicle in the grid path and time nodes corresponding to different grids in the time window; The adjustment module is used to adjust the time nodes when the time windows of different automated guided vehicles overlap or conflict, and re-plan the automated guided vehicles corresponding to the overlapping or conflicting time windows.

[0013] Optionally, the construction module is further configured to obtain a plane figure of the target area; determine the linear moving distance of the automated guided vehicle per unit time as the grid side length, divide the plane figure into a plurality of grids, and obtain the grid model.

[0014] Optionally, the planning module is further configured to determine the probability rule that the kth ant moves from point i to point j at time t. for: ; in is the concentration of pheromone, is the heuristic function, α is the pheromone heuristic factor, β is the expected heuristic factor, represents all possible locations that ant k may pass through when moving, i and j represent two different locations respectively; Introducing a direction function , the specific formula is as follows: ; The improved ant movement probability rule is as follows: ; in, is the angle between the direction from the starting point to the end point and the horizontal axis in the coordinate system constructed based on the grid model, is the angle between the ant from point i to point j and the horizontal axis, is the angle between the ant’s current forward direction and the direction from the starting point to the end point, The values are the angle parameter and the direction parameter respectively.

[0015] Optionally, the time window model can be expressed as: ; in, For any AGV, specify the time window corresponding to the path in the grid model. is the time it takes for the AGV to pass through any grid in the specified path, It is the time point when the AGV head enters any of the grids. It is the time point when the tail of the AGV leaves any grid.

[0016] Optionally, the adjustment module is also used to determine the priority of the automated guided vehicle corresponding to the overlapping or conflicting time window; adjust the time window for the automated guided vehicle with a lower priority so that it avoids the overlapping or conflicting time window; or re-plan the grid corresponding to the overlapping or conflicting time node so that the automated guided vehicle with a lower priority detours or waits.

[0017] Optionally, the adjustment module is further configured to determine the priority of the automated guided vehicle according to the mission urgency, grid path efficiency, and current operating status of the automated guided vehicle.

[0018] The present invention also provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, the above-mentioned automatic guided vehicle collision avoidance method is implemented.

[0019] The present invention also provides a computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the above-mentioned automatic guided vehicle collision avoidance method when executing the program.

[0020] The automatic guided vehicle collision avoidance method provided by the present invention has the following beneficial effects: By introducing the direction function to optimize the ant colony algorithm, the initial search efficiency and convergence speed of the ant colony algorithm are improved. In this way, the path planning is first performed through the improved ant colony algorithm, which can comprehensively consider the environmental map and AGV status information to obtain the global optimal solution. In addition, the direction function, as a new heuristic method, provides clearer search direction guidance and considers the path directionality at the same time when selecting the path, thereby achieving more targeted search and improving the planning efficiency of the grid path. Combined with the time window, the AGV path can be dynamically adjusted according to the real-time environment and traffic conditions, and respond to emergencies or environmental changes in real time, thereby achieving more flexible and efficient collision avoidance planning to adapt to the ever-changing environment and AGV status, reducing delays and downtime. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] To more clearly illustrate the embodiments of the present invention and its design, the following briefly introduces the drawings required for this embodiment. The drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be derived from these drawings without inventive effort.

[0022] Figure 1 The present invention is a flowchart of an automated guided vehicle collision avoidance method according to an exemplary embodiment of the present invention.

[0023] Figure 2 The present invention is a block diagram of an automatic guided vehicle collision avoidance device according to an exemplary embodiment. DETAILED DESCRIPTION

[0024] In order to enable those skilled in the art to better understand the technical solution of the present invention and to be able to implement it, the present invention is described in detail below with reference to the accompanying drawings and specific embodiments. The following embodiments are only used to more clearly illustrate the technical solution of the present invention and are not intended to limit the scope of protection of the present invention.

[0025] To address the problem that existing AGV path planning methods cannot quickly obtain an optimal solution, this paper proposes a fusion ant colony algorithm that combines the ant colony algorithm with the time window principle. First, in a large factory environment, this method uses the ant colony algorithm after optimizing the direction function to perform path planning and generate the optimal path for multiple AGVs. Second, using the time window principle, the time window of each AGV passing through the node is calculated and the time window table of each node is updated. Finally, each node is compared to see whether there is a time window conflict. If there is no conflict, the next step is continued. If there is a conflict, the conflicting nodes are added to the taboo table and compared again after resolving the time window conflict. If a time window conflict with an adjacent node occurs simultaneously during the path planning process, resulting in the inability to execute the task, the problem is resolved by expanding the search range of the time node and repeating a new round of search. This completes multi-AGV path planning, solves the multi-AGV collision avoidance problem, and improves system flexibility and efficiency.

[0026] The technical solutions provided by various embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0027] First, the present invention provides a method for avoiding collision of an automatic guided vehicle, specifically Figure 1 As shown, the following steps are included: S101: Construct a grid model of the target area.

[0028] Specifically, first, a plane figure of the target area is obtained; secondly, the linear moving distance of the automated guided vehicle per unit time is determined as the grid side length, and the plane figure is divided into multiple grids to obtain the grid model.

[0029] For example, a grid method is used to simplify the environment modeling. Specifically, the target area, or the actual operating environment, is represented as a two-dimensional plane. This plane is called a grid map. This map contains many grids of unit size. These grids are divided into two categories: obstacle grids and traversable grids. Obstacle grids represent impassable areas, while traversable grids represent freely traversable areas. They are represented on the map in black and white, respectively.

[0030] In the above grid environment, the characteristics of each AGV are assumed to be as follows: (1) Each AGV occupies a grid, and its unit moving speed is equal to the side length of the grid.

[0031] (2) All AGVs have the same speed, whether empty or loaded. Turning time is constant.

[0032] (3) Ensure that the size of the unit feasible grid is sufficient for the AGV to pass smoothly, so as to avoid the grid size affecting the AGV path planning.

[0033] (4) Each AGV can only accept and complete one task at a time, and can only execute the next task after the task is completed.

[0034] S102. Introducing a direction function into the ant colony algorithm to obtain an improved ant colony algorithm, specifying a direction of a path using the improved ant colony algorithm, and performing path planning on the automatic guided vehicle on the grid model based on the specified direction to obtain a grid path.

[0035] In this step, an ant colony algorithm is used to plan the AGV's path. Initially, ants are simulated releasing pheromones within a virtual graphical environment. These pheromones accumulate along the edges of the graph, indicating the ants' previous movements. As ant colony members randomly explore the graph, they select paths based on pheromone concentrations, favoring them over more optimal paths. The pheromones on these paths gradually evaporate over time, preventing the algorithm from prematurely falling into local optima and neglecting other potentially more optimal paths. After multiple iterations, the algorithm gradually stabilizes and finds the most cost-effective path. Ultimately, this path is selected as the AGV's route.

[0036] Specifically, the application of traditional ant colony algorithms to AGV path planning is mainly divided into foraging rules, pheromone rules, and movement rules. The foraging rule requires m ants to crawl from a specified starting point to an end point, and a taboo table is set to specify obstacles (areas where ants cannot travel). The pheromone rule stipulates that the m ants will leave a certain amount of pheromone along the path from the starting point to the end point, and the pheromone evaporates over time. The movement rule stipulates that the probability of ant k moving from the current node i to the next node j is determined by the pheromone concentration and a heuristic function.

[0037] The probability rule that the kth ant moves from point i to point j at time t for: ; is the concentration of pheromone, is the heuristic function, α is the pheromone heuristic factor, β is the expected heuristic factor, represents all possible locations that ant k may pass through when moving, i and j represent two different locations respectively; the heuristic function is the Euler distance from node i to node j The reciprocal of , that is: ; When all ants have completed their search, the pheromones on the path are updated and the rules are updated. for: ; Where, is the volatility coefficient, It represents the increment of pheromone on the path, and the calculation formula is: ; Where Q is a constant, is the set ant pheromone intensity, is the total length of the path traversed by ant k in one cycle.

[0038] Traditional ant colony algorithms have disadvantages in AGV path planning, such as low initial search efficiency and slow convergence. Furthermore, the ant colony algorithm is a global search, and the ants are aimless and directionless when searching for a path from the starting point to the end point. To address this issue, the present invention makes the following improvements:

[0039] Introducing a direction function , the specific formula is as follows: ; The improved ant movement probability rule is as follows: ; in It is the angle between the direction from the starting point to the end point and the horizontal axis in the coordinate system constructed based on the grid model. It is the angle between the ant from point i to point j and the horizontal axis. Generally, the ant moves in eight directions for the next step, namely {0, 45, 90, 135, 180, 225, 270, 325}, excluding the seven directions before point i. is the angle between the ant’s current forward direction and the direction from the starting point to the end point, and The closer the difference is to 0, The closer it is to 1, the closer the ant's moving direction is to the starting and ending points. If the difference is too large, other values are assigned. The values are angle parameters and direction parameters, which should be appropriately selected according to the specific properties of the path.

[0040] S103: Determine a time window of each automated guided vehicle in the grid path and time nodes corresponding to different grids in the time window.

[0041] In this step, the time windows for all AGVs passing through each node are calculated and the time window table for each node is updated. Next, the time window tables for all nodes are updated. During this process, each node is compared to see if a time window conflict occurs. If no conflict occurs, the process proceeds to the next step. If a conflict does occur, the conflicting nodes are added to the conflict table to resolve the time window conflict and compared again.

[0042] For example, to prevent conflicts and deadlocks in a two-way, single-lane AGV system, it is necessary to clearly define the specific time period each AGV occupies on a road segment. This time period is also known as the time window for AGV system path planning. The core meaning of the time window is the period from when an AGV enters a path to when it leaves the path. During this time period, the vehicle has the right to travel along a certain direction along the path. In AGV path planning, the time window algorithm is widely used due to its simplicity and practicality. The time window can be expressed as:

[0043] ; in, For any AGV, specify the time window corresponding to the path in the grid model. is the time it takes for the AGV to pass through any grid in the specified path. It is the time point when the AGV head enters any grid. It is the time point when the tail of the AGV leaves any grid.

[0044] In addition, after finding the preliminary path, the path can be smoothed to reduce unnecessary turns and sudden stops, thereby improving the driving efficiency and safety of the AGV.

[0045] S104: When the time windows of different automated guided vehicles overlap or conflict, adjust the time node and replan the automated guided vehicles corresponding to the overlapping or conflicting time windows.

[0046] In this step, if a time window conflict with an adjacent node during path planning prevents the task from executing, the solution is to adjust the search range for the time nodes. This means adding new nodes or time ranges to the original search to find a feasible path. The search is then repeated to ensure successful task execution.

[0047] For example, during the AGV system's path planning process, if a conflict with a neighboring node in a time window prevents the task from being executed, this problem can be resolved by adjusting the search range of the time window. Specifically, the start and end times of the time window are adjusted to re-evaluate the AGV's start and end times and find a new conflict-free time window. A new round of path search then occurs, which involves re-evaluating the driving route, including possible detours, to ensure that conflicts with other AGVs are avoided within the adjusted time window. Furthermore, the system optimizes path selection, reducing overall travel time and improving efficiency.

[0048] For example, conflicts can be identified first, allowing AGVs to be monitored and detected. The system monitors the AGV's position, speed, and direction of travel, as well as their relative positions, in real time. A time window model is used to detect any temporal overlap or spatial conflicts. Conflicts are then classified based on type (e.g., co-location conflicts, opposing conflicts), allowing for different resolution strategies. Furthermore, real-time data (e.g., new task requests, route changes, AGV status updates) can be used to dynamically update the route plan. Subsequent route planning can be adjusted based on feedback from the AGV's actual driving conditions (e.g., speed changes, route deviations, etc.).

[0049] For another example, when replanning a path, time windows can be dynamically adjusted. Specifically, priority ranking can be performed: First, the priority of the AGV is determined based on its mission urgency, grid path efficiency, and current operating status. After determining the priority of the AGV corresponding to the overlapping or conflicting time window, the time window is expanded: the time window of the lower-priority AGV is adjusted to avoid conflict with the higher-priority AGV, allowing it to avoid the overlapping or conflicting time window. This can be achieved by delaying or advancing its start and end time. Alternatively, the grid corresponding to the overlapping or conflicting time node for the lower-priority AGV can be replanned, forcing it to detour through the overlapping path or stop and wait to avoid the conflicting time node. Furthermore, heuristic information can be introduced into the ant colony algorithm to improve planning efficiency: heuristic information such as path length, number of turns, and obstacle density can be introduced into the ant colony algorithm to guide the ants' search direction. This information can help the algorithm find high-quality paths more quickly.

[0050] In one possible implementation, the automatic guided vehicle collision avoidance method proposed in the present invention can be implemented through the following steps.

[0051] Step 1: The large factory environment is modeled as a grid system using a grid method. This simplifies the complex factory layout into a grid composed of multiple cells. The grid method helps determine the paths that the AGV can pass through. Each grid represents an area within the factory, helping to convert continuous space into discrete units, thereby simplifying the complexity of path planning. White grids are passable grids, black grids are obstacle grids, blue grids are waiting areas, green grids are charging areas, and yellow grids are placement areas.

[0052] Step 2: Use the ant colony algorithm to implement preliminary AGV path planning. First, set the iteration counter G to 0. Then, place M ants at the starting position and add the starting point to the taboo table. Next, calculate the probability of each ant k being able to transfer to all nodes based on the transfer probability formula, and select the next node based on the roulette rule. During this process, the taboo table is continuously updated, and the paths found by all ants and their lengths are stored. If an ant "dies", that is, it cannot move forward, its path length is assumed to be infinite. Next, update the global pheromone according to the pheromone update formula, and clear the taboo table. Subsequently, determine whether the iteration counter G has reached the maximum number of iterations. If so, the process ends, and the optimal path for the AGV is determined. If not, repeat the above steps until the optimal path is found.

[0053] Step 3: Use a time window algorithm to detect multi-AGV collisions. Each AGV is assigned a precise time window for each step, defining the time period during which the AGV occupies the grid. This allows for precise control of each AGV's movement throughout the transport process, ensuring system coordination and precise timing. The system comprehensively calculates the time windows for all AGVs on their respective paths to prevent potential conflicts or delays along the routes and ensure that their actions do not overlap, leading to decreased efficiency. If overlapping time windows are detected, indicating an impending AGV collision, the system proceeds to Step 4. If no time window overlap occurs, the system proceeds to Step 5.

[0054] Step 4: When a potential collision or conflict is detected, dynamic rerouting is performed based on the priority of each AGV. The priority of each AGV is determined by its mission urgency, grid path efficiency, and current operational status. The rerouting is then recalculated to ensure all AGVs complete their missions in the optimal time and safety. Upon completion, the system proceeds to Step 5.

[0055] In the above steps, the priority of AGV can be determined specifically through the following aspects.

[0056] Task urgency assessment, for example, the remaining time window: calculates the difference between the time required for the AGV to complete the task and the remaining time from the current time to the task deadline. The smaller the difference, the higher the urgency of the task; task importance: assign different weights to different types of tasks. Urgent material delivery or production line fault repair may have a higher priority.

[0057] Predicted route efficiency analysis, such as route length and duration, assesses the expected travel time and distance for different routes. Shorter routes or faster travel speeds are considered more efficient. Route congestion analysis considers other AGVs or obstacles along the route and how these factors affect route efficiency. Use real-time data or predictive models to assess current and expected route congestion levels.

[0058] The current operating status of the AGV, for example, battery charge: AGVs with low battery may need priority access to charging stations to avoid affecting overall mission execution due to depletion of power; maintenance status: check whether the AGV needs regular maintenance or repairs, which may affect its ability to perform tasks; load condition: consider whether the AGV is currently carrying an important or urgent load, which may increase its priority.

[0059] For example, a comprehensive scoring model can be designed to prioritize each AGV by combining the three aforementioned evaluations. For example, a weight can be assigned to each factor (adjusted based on business needs and actual conditions) and then a total score can be calculated for each AGV.

[0060] Dynamic adjustment, for example, real-time updates: The system should be able to receive and process new data in real time (such as new task requests, path changes, AGV status updates, etc.), and dynamically adjust priorities accordingly; when multiple AGVs have similar priorities, more specific strategies can be used to resolve conflicts, such as selecting based on the shortest avoidance path, minimum waiting time, or minimum impact range.

[0061] User intervention, for example, may also need to allow operators or system administrators to intervene manually in certain situations, especially in extreme or special circumstances, to ensure the flexibility and responsiveness of the system.

[0062] Through the above method, it can be ensured that when a potential collision or conflict is detected, the system can quickly and reasonably replan the path and assign priorities to AGVs, thereby maximizing overall efficiency and safety.

[0063] Step 5: Output a set of conflict-free time windows and routing arrangements. This arrangement ensures that all AGVs can operate in a coordinated manner across different temporal and spatial dimensions. This highly coordinated operation not only improves the overall efficiency of the AGV system, reducing waiting and idle time, but also significantly enhances safety, ensuring continuous production flow and logistics efficiency. Through this meticulous management, the system maintains stable and reliable performance even as the number of AGVs increases and the tasks become more complex.

[0064] The above method is used to optimize the ant colony algorithm by introducing a direction function, which improves the initial search efficiency and convergence speed of the ant colony algorithm. The direction function, as a new heuristic method, provides clearer search direction guidance and considers the path directionality during path selection, thereby achieving more targeted search and improving the planning efficiency of the grid path. Combined with the time window, the AGV path can be dynamically adjusted according to the real-time environment and traffic conditions, and respond to emergencies or environmental changes in real time, thereby achieving more flexible and efficient collision avoidance planning to adapt to the changing environment and AGV status, reducing delays and downtime.

[0065] Secondly, the present invention also provides an automatic guided vehicle collision avoidance device, such as Figure 2 Shown, including: The construction module 201 is used to construct a grid model of the target area.

[0066] The planning module 202 is used to introduce a direction function into the ant colony algorithm to obtain an improved ant colony algorithm, specify the direction of the path through the improved ant colony algorithm, and perform path planning on the automatic guided vehicle on the grid model based on the specified direction to obtain a grid path.

[0067] The determination module 203 is configured to determine the time window of each automated guided vehicle in the grid path and the time nodes corresponding to different grids in the time window.

[0068] The adjustment module 204 is configured to adjust the time node when the time windows of different automated guided vehicles overlap or conflict, and replan the automated guided vehicles corresponding to the overlapping or conflicting time windows.

[0069] Optionally, the construction module 201 is further configured to obtain a plane figure of the target area; determine the linear moving distance of the automated guided vehicle per unit time as the grid side length, divide the plane figure into a plurality of grids, and obtain the grid model.

[0070] Optionally, the planning module 202 is further configured to determine the probability rule that the kth ant moves from point i to point j at time t. for: ; in is the concentration of pheromone, is the heuristic function, α is the pheromone heuristic factor, β is the expected heuristic factor, Represents all the possible locations that ant k may pass through when moving, i and j represent two different locations respectively; introduce a direction function , the specific formula is as follows: ; The improved ant movement probability rule is as follows: ; in, It is the angle between the direction from the starting point to the end point and the horizontal axis in the coordinate system constructed based on the grid model. is the angle between the ant from point i to point j and the horizontal axis, is the angle between the ant’s current forward direction and the direction from the starting point to the end point, The values are the angle parameter and the direction parameter respectively.

[0071] Optionally, the time window model can be expressed as: ; in, For any AGV, specify the time window corresponding to the path in the grid model. is the time it takes for the AGV to pass through any grid in the specified path. It is the time point when the AGV head enters any grid. It is the time point when the tail of the AGV leaves any grid.

[0072] Optionally, the adjustment module 204 is also used to determine the priority of the automated guided vehicle corresponding to the overlapping or conflicting time window; adjust the time window for the automated guided vehicle with a lower priority so that it avoids the overlapping or conflicting time window; or re-plan the grid corresponding to the overlapping or conflicting time node so that the automated guided vehicle with a lower priority detours or waits.

[0073] Optionally, the adjustment module 204 is further configured to determine the priority of the automated guided vehicle according to the mission urgency, grid path efficiency, and current operating status of the automated guided vehicle.

[0074] The present invention also provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, the above-mentioned automatic guided vehicle collision avoidance method is implemented.

[0075] The present invention also provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, the above-mentioned automatic guided vehicle collision avoidance method is implemented.

[0076] By adopting the above-mentioned device, the ant colony algorithm is optimized by introducing a direction function, which improves the initial search efficiency and convergence speed of the ant colony algorithm. The direction function, as a new heuristic method, provides clearer search direction guidance and considers the path directionality at the same time when selecting the path, thereby achieving a more targeted search and improving the planning efficiency of the grid path. In combination with the time window, the AGV path can be dynamically adjusted according to the real-time environment and traffic conditions, and respond to emergencies or environmental changes in real time, thereby achieving more flexible and efficient collision avoidance planning to adapt to the ever-changing environment and AGV status, reducing delays and downtime.

[0077] The present invention also provides a computer-readable storage medium, which stores a computer program, which can be used to execute the above Figure 1 Provided is a collision avoidance method for an automated guided vehicle.

[0078] The present invention also provides a computer device. At the hardware level, the computer device includes a processor, an internal bus, a network interface, a memory, and a non-volatile memory. Of course, it may also include hardware required for other services. The processor reads the corresponding computer program from the non-volatile memory into the memory and then runs it to achieve the above Figure 1 Provided is a collision avoidance method for an automated guided vehicle.

[0079] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Thus, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0080] The present invention is described with reference to flowcharts and / or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowcharts and / or block diagrams, as well as combinations of processes and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowcharts and / or block diagrams. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0081] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0082] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0083] It should be noted that the above specific embodiments can enable those skilled in the art to more fully understand the present invention, but do not limit the present invention in any way. Therefore, although this specification has described the present invention in detail, those skilled in the art should understand that the present invention can still be modified or replaced with equivalents; and all technical solutions and improvements that do not depart from the spirit and scope of the present invention are included in the scope of protection of the patent for the present invention. Any reference signs in the claims should not be construed as limiting the claims involved.

Claims

1. A collision avoidance method for an automated guided vehicle, characterized in that: The method comprises: Construct a grid model of the target area; A direction function is introduced into the ant colony algorithm to obtain an improved ant colony algorithm, a direction of a path is specified by the improved ant colony algorithm, and a path is planned for the automatic guided vehicle on the grid model based on the specified direction to obtain a grid path; Determining a time window of each automated guided vehicle in the grid path and time nodes corresponding to different grids in the time window; In the case where the time windows of different automated guided vehicles overlap or conflict, the time nodes are adjusted and the automated guided vehicles corresponding to the overlapping or conflicting time windows are replanned.

2. The automatic guided vehicle collision avoidance method according to claim 1, characterized in that: The grid model of the target area is constructed as follows: Obtain the plane graphics of the target area; The linear moving distance per unit time of the automatic guided vehicle is determined as the side length of the grid, and the plane figure is divided into a plurality of grids to obtain the grid model.

3. The method for avoiding collision of an automated guided vehicle according to claim 1, wherein: The direction function is introduced into the ant colony algorithm, and the improved ant colony algorithm includes: Determine the probability rule that the kth ant moves from point i to point j at time t for: ； in is the concentration of pheromone, is the heuristic function, α is the pheromone heuristic factor, β is the expected heuristic factor, represents all possible locations that ant k may pass through when moving, i and j represent two different locations respectively; Introducing a direction function , the specific formula is as follows: ； The improved ant movement probability rule is as follows: ； in, is the angle between the direction from the starting point to the end point and the horizontal axis in the coordinate system constructed based on the grid model, is the angle between the ant from point i to point j and the horizontal axis, is the angle between the ant’s current forward direction and the direction from the starting point to the end point, The values are the angle parameter and the direction parameter respectively.

4. The method for avoiding collision of an automated guided vehicle according to claim 1, wherein: The time window is expressed as: ； in, For any AGV, specify the time window corresponding to the path in the grid model. is the time it takes for the AGV to pass through any grid in the specified path, It is the time point when the AGV head enters any of the grids. It is the time point when the tail of the AGV leaves any grid.

5. The method for avoiding collision of an automated guided vehicle according to claim 1, wherein: Adjusting the time nodes and replanning the automated guided vehicles corresponding to the overlapping or conflicting time windows includes: Determine the priority of AGVs corresponding to overlapping or conflicting time windows; Adjust the time window for the automated guided vehicle with lower priority to avoid the overlapping or conflicting time window; or replan the grid corresponding to the overlapping or conflicting time node to make the automated guided vehicle with lower priority detour or wait.

6. The method for avoiding collision of an automated guided vehicle according to claim 5, characterized in that: The priority of the automated guided vehicle is determined by the mission urgency, grid path efficiency and current operating status of the automated guided vehicle.

7. An automatic guided vehicle collision avoidance device, characterized in that: The device comprises: A construction module is used to construct a grid model of the target area; a planning module, configured to introduce a direction function into the ant colony algorithm to obtain an improved ant colony algorithm, specify a path direction using the improved ant colony algorithm, and perform path planning for the automated guided vehicle on the grid model based on the specified direction to obtain a grid path; a determination module, configured to determine a time window of each automated guided vehicle in the grid path and time nodes corresponding to different grids in the time window; The adjustment module is used to adjust the time nodes when the time windows of different automated guided vehicles overlap or conflict, and re-plan the automated guided vehicles corresponding to the overlapping or conflicting time windows.

8. A computer-readable storage medium, characterized in that The storage medium stores a computer program, and when the computer program is executed by the processor, the automatic guided vehicle collision avoidance method according to any one of claims 1 to 6 is implemented.

9. A computer device, characterized in that: The method comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the method for avoiding collision of an automated guided vehicle as claimed in any one of claims 1 to 6 is implemented.

Images