Twin factory AGV path planning method, system, medium and equipment

Abstract

The application discloses a kind of twin factory AGV path planning method, system, medium and equipment, it is related to path planning field, including obtaining and constructing twin factory 3D model;With the shortest path planning time and the minimum planning path length as the objective function of path planning, the PSO-A* algorithm is used to optimize the path, to determine the global optimal path planning result of automatic guided vehicle AGV in twin factory 3D model;Wherein, the PSO-A* algorithm is improved by introducing PSO algorithm to iteratively update the weight of heuristic function in evaluation function in A algorithm, to carry out path planning, including: determining the initial fitness value of each particle based on objective function, using particle swarm optimization PSO algorithm to iteratively update the position and speed of each particle in particle swarm, determine the fitness value of each particle after iteration update, the weight value corresponding to the particle position of optimal fitness value is used as the new weight of heuristic function in evaluation function in improved A* algorithm.

Description

Technical Field

[0001] This invention relates to the field of path planning, and in particular to a path planning method, system, medium, and equipment for AGVs in a twin factory. Background Technology

[0002] With the popularization of the concept of smart manufacturing, automated and intelligent production systems are playing an increasingly important role in modern manufacturing. AGVs, as a key component of smart factories, are widely used in tasks such as material handling and parts delivery to improve production efficiency, reduce labor costs, and improve the working environment. AGV path planning is a crucial process in smart manufacturing, directly affecting production efficiency, logistics speed, and the level of factory automation. Efficient path planning not only ensures the timely and accurate delivery of materials and parts but also reduces handling time, optimizes production processes, and improves factory efficiency.

[0003] Path planning is a core problem in robotics and automation systems, especially important in the application of Automated Guided Vehicles (AGVs) in smart factories. The goal of path planning is to find the optimal or near-optimal path from the starting point to the destination in a given environment, while satisfying certain constraints, such as avoiding obstacles and minimizing travel time or distance.

[0004] Existing technologies typically use the traditional A* algorithm for AGV path optimization. However, in practical applications, path planning is often time-consuming. For example, the planned path may contain a large number of redundant nodes, which requires frequent adjustments to the posture during actual movement, resulting in low efficiency of AGV in transporting goods. Summary of the Invention

[0005] This invention provides a twin factory AGV path planning method, system, medium, and equipment to solve the aforementioned problems existing in the prior art, namely, how to achieve efficient path planning in the prior art. This invention provides a twin factory AGV path planning method, which includes: Obtain map data of the factory to be planned and construct a 3D twin factory model including automated guided vehicles (AGVs); The objective function of path planning is to minimize path planning time and path length. Based on map data and the objective function, the PSO-A* algorithm is used for path optimization to determine the globally optimal path planning result for the Automated Guided Vehicle (AGV) in the 3D twin factory model. The PSO-A* algorithm iteratively updates the weights of the heuristic function in the evaluation function of the improved A* algorithm by introducing a Particle Swarm Optimization (PSO) algorithm. This includes: initializing particles, assigning a weight to the heuristic function in the evaluation function for each particle; determining the initial fitness value of each particle based on the objective function; iteratively updating the position and velocity of each particle in the swarm using the PSO algorithm; determining the fitness value of each particle after the iterative update; using the weight value corresponding to the position of the particle with the optimal fitness value as the new weight of the heuristic function in the evaluation function of the improved A* algorithm; and outputting the corresponding path planning result. The improved A* algorithm optimizes the penalty function used to adjust the AGV's direction change in the cost function of the original A* algorithm based on a dynamic penalty mechanism. The velocity represents the direction and distance the particle moves in the next iteration.

[0006] Optionally, the Manhattan distance is selected as the heuristic function, and the acquisition of the heuristic function specifically includes: ; in, For heuristic functions, For Manhattan distance, The x-coordinate of the current node. The y-coordinate of the current node. Let x be the x-coordinate of the target node. y is the ordinate of the target node.

[0007] Optionally, obtaining the evaluation function specifically includes: ; in, Let k be the evaluation function, and k be the weight of the heuristic function of the A* algorithm. This represents the actual cumulative cost from the starting point to node n. This is a heuristic function.

[0008] Optionally, obtaining the objective function specifically includes: ; in, Let be the objective function. This represents the time required to plan the route. The path length is the planned path. and The weight values ​​corresponding to the time required to plan the path and the weight values ​​corresponding to the path length of the planned path are respectively assigned.

[0009] Optionally, obtaining the cost function specifically includes: ; in, Let cost function be The actual cumulative cost from the starting point to n. Based on the cost of movement and =10, For the penalty function; in, ; in, The Manhattan distance from the starting point to the destination. As a penalty factor, For the intensity of punishment and .

[0010] Optionally, YOLOv8n can be used to detect and obtain the QR code of the automated guided vehicle (AGV) and upload it to the cloud server to synchronize the location information of the AGV.

[0011] This invention provides a twin factory AGV path planning system, comprising: The building module is used to acquire map data of the factory to be planned and build a 3D twin factory model including automated guided vehicles (AGVs). The path planning module uses the shortest path planning time and the minimum planned path length as the objective function. Based on map data and the objective function, it employs the PSO-A* algorithm for path optimization to determine the globally optimal path planning result for the Automated Guided Vehicle (AGV) in the twin factory 3D model. The PSO-A* algorithm iteratively updates the weights of the heuristic function in the evaluation function of the improved A* algorithm by introducing a Particle Swarm Optimization (PSO) algorithm. This includes: initializing particles, assigning a weight to the heuristic function in the evaluation function for each particle; determining the initial fitness value of each particle based on the objective function; iteratively updating the position and velocity of each particle in the swarm using the PSO algorithm; determining the fitness value of each particle after the iterative update; using the weight value corresponding to the position of the particle with the optimal fitness value as the new weight of the heuristic function in the evaluation function of the improved A* algorithm; and outputting the corresponding path planning result. The improved A* algorithm optimizes the penalty function used to adjust the AGV's direction change in the cost function of the original A* algorithm based on a dynamic penalty mechanism. The velocity represents the direction and distance the particle moves in the next iteration.

[0012] The present invention provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described twin factory AGV path planning method.

[0013] The present invention provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the above-described twin factory AGV path planning method.

[0014] Compared with existing technologies, the beneficial effects of this invention are as follows: This invention provides a twin factory AGV path planning method. This method finds the optimal heuristic function weight by using the global search capability of the PSO algorithm, avoiding the A* algorithm from getting trapped in local optima and significantly improving pathfinding efficiency. By introducing a dynamic penalty mechanism, unnecessary inflection points in the path are effectively reduced, improving path quality. In addition, by dynamically adjusting the penalty function, this invention avoids the suboptimal path selection problem caused by fixed penalties, ensuring the optimality or suboptimality of the path and effectively reducing the time required for path planning. It can achieve efficient and high-quality path planning, reduce the time required for AGV logistics transportation, and effectively improve the production level and efficiency of the factory. Attached Figure Description

[0015] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.

[0016] Figure 1 A flowchart of a twin factory AGV path planning method provided in an embodiment of the present invention; Figure 2 A schematic diagram illustrating the construction process of the PSO-A* algorithm provided in an embodiment of the present invention; Figure 3 This is a schematic diagram illustrating the performance comparison of algorithms under a simplified graph structure, as provided in an embodiment of the present invention. Figure 4 This is a schematic diagram comparing the algorithm performance under complex map structures provided in the embodiments of the present invention; Figure 5 The figure shows the experimental results of the twin warehouse verification provided in the embodiments of the present invention; Figure 6 A schematic diagram of a computer device for the twin factory AGV path planning method provided in an embodiment of the present invention. Detailed Implementation

[0017] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0018] The technical solution of the present invention and how the technical solution of the present invention solves the above-mentioned technical problems are described in detail below with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments. The embodiments of the present invention will now be described with reference to the accompanying drawings.

[0019] Example 1 Figure 1 This is a flowchart of a twin factory AGV path planning method provided by an embodiment of the present invention, such as... Figure 1 As shown in the figure, this embodiment illustrates a twin factory AGV path planning method, including: S1: Obtain map data of the factory to be planned and construct a 3D twin factory model including automated guided vehicles (AGVs).

[0020] For example, a 3D model of the factory can be created by acquiring a map of the indoor scene, and a digital twin system of the factory can be built based on data collected from sensing devices installed in the physical warehouse. Sensors and radio frequency identification tags (RFID) suitable for the factory environment can be selected and installed in appropriate locations in the factory. RFID tags can be attached to materials or parts, each containing a unique identification number and related information. RFID readers can scan and read the information on the RFID tags, transmitting the information data sensed by the sensors and read by the RFID readers, such as the operating status and speed of AGVs, and the names, batch numbers, and production dates of parts, to the digital twin factory.

[0021] For example, IoT technology can be used to collect data related to the physical product warehouse using various sensors, including data on temperature and humidity, inventory information, and equipment status. The data is then preprocessed and analyzed before being transmitted to the twin factory model.

[0022] S2: The objective function of path planning is to minimize the path planning time and the planned path length. Based on map data and the objective function, the PSO-A* algorithm is used for path optimization to determine the globally optimal path planning result of the Automated Guided Vehicle (AGV) in the twin factory 3D model. The PSO-A* algorithm iteratively updates the weights of the heuristic function in the evaluation function of the improved A* algorithm by introducing the Particle Swarm Optimization (PSO) algorithm. This includes: initializing particles, assigning a weight to the heuristic function in the evaluation function for each particle; determining the initial fitness value of each particle based on the objective function; iteratively updating the position and velocity of each particle in the swarm using the PSO algorithm; determining the fitness value of each particle after the iterative update; using the weight value corresponding to the position of the particle with the optimal fitness value as the new weight of the heuristic function in the evaluation function of the improved A* algorithm; and outputting the corresponding path planning result. The improved A* algorithm optimizes the penalty function used to adjust the AGV's direction change in the cost function of the original A* algorithm based on a dynamic penalty mechanism. The velocity represents the direction and distance the particle moves in the next iteration.

[0023] For example, the construction and optimization process of the PSO-A* algorithm proposed in this invention may specifically include: (1) Constructing the heuristic function for the A* algorithm: Heuristic function h ( n The Manhattan distance is chosen; therefore, the heuristic function can be described as: ; in, Let be a heuristic function, representing the heuristically estimated cost from the current node (starting point) to the target node (ending point). For Manhattan distance, The x-coordinate of the current node. The y-coordinate of the current node. Let x be the x-coordinate of the target node. y is the ordinate of the target node.

[0024] Heuristic function weights assigned to the A* algorithm Therefore, the evaluation function of the A* algorithm is: ; in, Let k be the evaluation function, and k be the weight of the heuristic function of the A* algorithm. This represents the actual cumulative cost from the starting point to node n.

[0025] (2) Optimization of the heuristic function of the A* algorithm in the offline stage: The weight k of the heuristic function of the A* algorithm is optimized using the PSO algorithm. The process is as follows: Step 2.1: Combine map data and Import into the PSO optimization program; Step 2.2: Define the parameter search space. Since the objective function considers both time and path length, the particle's position (weight) The search range for the value is set to (0.1, 5).

[0026] Step 2.3: Initialize the particle swarm. Each particle in the particle swarm represents... Medium weight An initial value is a potential value that can be taken from the search space. These initial values ​​are uniformly distributed throughout the search space to prevent the algorithm from getting trapped in local optima.

[0027] Step 2.4: Calculate the fitness value. This involves assigning a fitness value to each particle. Substitute the value into the evaluation function In the process, the planned path time is obtained after running the A* algorithm. and path length Substitute into the objective function of the formula The fitness value of the particle is calculated based on... Evaluate the performance of the A* algorithm. The objective function can be described as follows:

[0028] ; in, This represents the time required to plan the route. The path length is the planned path. and The weight value and = =0.5, when > When optimizing for shorter time, it may lead to a suboptimal path. < Optimizing towards shorter paths may increase planning time.

[0029] Step 2.5: Update the individual optimal solution and global optimal solution for each particle. Set the current fitness value of each particle... The particle's fitness value is compared to its previously recorded best fitness value. If the current fitness value is better, the particle's individual best solution is updated. After all particles have updated their individual best solutions, the entire particle swarm is traversed, and the individual best solutions of each particle are compared to find the particle with the best fitness value and its position (weight) is determined. The value is recorded as the global optimal solution. Velocity and position are updated according to the velocity and position update formulas, where the velocity... ,Location The update formula can be described as follows:

[0030] ; ; ; in, and It is a random number between (0, 1). and Pbest represents the learning factor. i gbest represents an individual extreme value. i Represents the global extremum. For the updated speed, The updated position These are weight values; Step 2.6: Calculate the fitness value of each particle after the update, compare the best fitness value of each particle with the fitness value at its historical best position, if it is better, then take its current position as the best position of the particle. For each particle, compare the fitness value corresponding to its best position with the best fitness value of the population, and update the best position and best fitness value of the population. Step 2.7: Determine if the search results meet the stopping condition. If they do, output the optimal value; otherwise, go to Step 2.4 and continue running until the condition is met to obtain the optimal weight. Value and save; The optimal weight k obtained by optimizing the PSO algorithm is assigned to the heuristic function of the A* algorithm to obtain the optimized PSO-A* algorithm.

[0031] For example, the specific process of introducing a dynamic penalty mechanism, which involves introducing a penalty function with changing direction into the cost function of the A* algorithm, is as follows: (1) The cost function of the A* algorithm A dynamic directional consistency penalty mechanism is introduced, which is applied when the A* algorithm expands from node n to neighboring node m, and its cost function... The calculation formula has been updated as follows: ; in, This represents the actual cumulative cost from the starting point to node n. Based on the cost of movement and =10, For the penalty function; (2) If the direction of movement changes, a penalty is imposed: ; in, The Manhattan distance from the starting point to the destination. As a penalty factor, For the intensity of punishment and , The size of the value will affect the quality of path finding. If it is too large, the path length will be sacrificed to reduce the number of inflection points, while if it is too small, the number of inflection points in the path will still be too large.

[0032] For example, to improve the efficiency of AGVs in task operations, the PSO-A* algorithm can be used to achieve efficient path planning for AGVs during task execution and obtain the optimal feasible path for AGVs.

[0033] For example, YOLOv8n can be used to detect and extract the QR code of the automated guided vehicle (AGV), and the pylibdmtx library can be used to decode the detected QR code. The decoded data is then transmitted to the cloud server via the HTTP protocol to synchronize the AGV's location information.

[0034] For example, the QR code recognition process may include the following steps: (1) Collect and create a Data Matrix QR code dataset, and use the dataset to train the YOLOv8n model; (2) Deploy the trained YOLOv8n model to an edge computing device; (3) The system captures the image to be detected in real time through the camera, and the edge computing device calls the deployed model to infer the captured image and locate the position of all QR codes in the image; (4) The system uses the pylibdmtx library to decode the location of the QR code and read the data information stored therein; (5) Transmit the successfully decoded data from the edge device to the remote cloud server via the HTTP protocol.

[0035] Example 2 For example, a simple 20x20 cell graph can be constructed in MATLAB (obstacles make up 20%). Each cell is represented by a unit of 1, black cells represent obstacles, and white grids represent passable areas. The starting point is set to (1, 1), and the ending point to (20, 20). Both the starting and ending points are represented by red cells, the path planned by the algorithm is represented by green lines, and nodes searched during the pathfinding process are represented by blue cells. Figure 3 (a) is the path generated using the traditional A* algorithm. Figure 3 (b) is the path generated by the PSO-A* algorithm after optimization using the PSO algorithm. Figure 3 (c) is the path generated after using the improved penalty strategy, which requires searching 79 nodes, reduces the number of inflection points by 6, and has a search time of 2.86 ms.

[0036] Construct a more complex map with 35×35 cells (obstacles make up 25% of the map), setting the starting point coordinates to (1, 1) and the ending point coordinates to (33, 33). Compare the results. Figure 4 (a) to Figure 4 (c); As shown in Table 1, the performance comparison before and after the algorithm improvement is presented.

[0037] Table 1 Performance comparison before and after algorithm improvement Compared to the traditional A* algorithm and the PSO-A* algorithm optimized by PSO in this invention, the improved penalized A* algorithm proposed in this invention achieves varying degrees of improvement in key performance indicators such as search time, number of search nodes, and number of inflection points in path planning. Specifically, search time is reduced by 60.44%-73.61%, the number of search nodes is reduced by 51.53%-71.88%, and the number of inflection points is reduced by 40%-47.06%.

[0038] For example, path planning for AGVs in a twin warehouse yields experimental results as follows: Figure 5 As shown in the image. The blue grid area represents the divided raster map, with a map size of 27×10m and a grid size of 1.15m×1.15m. The green lines represent the planned paths, the blue dots represent the starting points, and the red dots represent the target points. Figure 5 In (a), the starting point coordinates of the AGV are (2, 2), the target point coordinates are (17, 6), and the planned path is shown by the green line in the figure. Figure 5 In (b), the starting point coordinates of the AGV are (2, 2) and the target point coordinates are (17, 6). However, an obstacle appeared on the original path, and the planned new path is shown by the green line in the figure. Figure 5 In (c), the starting point coordinates of the AGV are (12, 7), the target point coordinates are (16, 1), and the planned path is shown by the green line in the figure.

[0039] Example 3 The above describes one or more embodiments of the twin factory AGV path planning method provided in this specification. Based on the same idea, this specification also provides a corresponding twin factory AGV path planning system, including: The building module is used to acquire map data of the factory to be planned and build a 3D twin factory model including automated guided vehicles (AGVs). The path planning module uses the shortest path planning time and the minimum planned path length as the objective function. Based on map data and the objective function, it employs the PSO-A* algorithm for path optimization to determine the globally optimal path planning result for the Automated Guided Vehicle (AGV) in the twin factory 3D model. The PSO-A* algorithm iteratively updates the weights of the heuristic function in the evaluation function of the improved A* algorithm by introducing a Particle Swarm Optimization (PSO) algorithm. This includes: initializing particles, assigning a weight to the heuristic function in the evaluation function for each particle; determining the initial fitness value of each particle based on the objective function; iteratively updating the position and velocity of each particle in the swarm using the PSO algorithm; determining the fitness value of each particle after the iterative update; using the weight value corresponding to the position of the particle with the optimal fitness value as the new weight of the heuristic function in the evaluation function of the improved A* algorithm; and outputting the corresponding path planning result. The improved A* algorithm optimizes the penalty function used to adjust the AGV's direction change in the cost function of the original A* algorithm based on a dynamic penalty mechanism. The velocity represents the direction and distance the particle moves in the next iteration.

[0040] Specific limitations regarding the twin factory AGV path planning system can be found in the limitations of the twin factory AGV path planning method described above, and will not be repeated here. Each module in the aforementioned twin factory AGV path planning system can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the memory of a computer device as software, so that the processor can call and execute the corresponding operations of each module.

[0041] Example 4 The present invention also provides a computer-readable storage medium storing a computer program that can be used to execute the twin factory AGV path planning method provided above.

[0042] Example 5 The present invention also provides Figure 6 The schematic diagram of the computer device shown is as follows: Figure 6 As shown, at the hardware level, the computer device includes a processor, internal bus, network interface, memory, and non-volatile memory, and may also include other hardware required for business operations. The processor reads the corresponding computer program from the non-volatile memory into memory and then runs it to implement the method provided in the above embodiments.

[0043] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the methods described above. Any references to memory, storage, databases, or other media used in the embodiments provided by this invention can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, or optical storage, etc. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc.

[0044] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this invention.

Claims

1. A twin factory AGV path planning method, characterized in that, include: Obtain map data of the factory to be planned and construct a 3D twin factory model including automated guided vehicles (AGVs); The objective function of path planning is to minimize path planning time and path length. Based on map data and the objective function, the PSO-A* algorithm is used for path optimization to determine the globally optimal path planning result for the Automated Guided Vehicle (AGV) in the 3D twin factory model. The PSO-A* algorithm iteratively updates the weights of the heuristic function in the evaluation function of the improved A* algorithm by introducing a Particle Swarm Optimization (PSO) algorithm. This includes: initializing particles, assigning a weight to the heuristic function in the evaluation function for each particle; determining the initial fitness value of each particle based on the objective function; iteratively updating the position and velocity of each particle in the swarm using the PSO algorithm; determining the fitness value of each particle after the iterative update; using the weight value corresponding to the position of the particle with the optimal fitness value as the new weight of the heuristic function in the evaluation function of the improved A* algorithm; and outputting the corresponding path planning result. The improved A* algorithm optimizes the penalty function used to adjust the AGV's direction change in the cost function of the original A* algorithm based on a dynamic penalty mechanism. The velocity represents the direction and distance the particle moves in the next iteration.

2. The twin factory AGV path planning method as described in claim 1, characterized in that, The Manhattan distance is selected as the heuristic function, and the acquisition of the heuristic function specifically includes: ； in, For heuristic functions, For Manhattan distance, The x-coordinate of the current node. The y-coordinate of the current node. Let x be the x-coordinate of the target node. y is the ordinate of the target node.

3. The twin factory AGV path planning method as described in claim 2, characterized in that, The acquisition of the evaluation function specifically includes: ； in, Let k be the evaluation function, and k be the weight of the heuristic function of the A* algorithm. This represents the actual cumulative cost from the starting point to node n. This is a heuristic function.

4. The twin factory AGV path planning method as described in claim 1, characterized in that, The acquisition of the objective function specifically includes: ； in, Let be the objective function. This represents the time required to plan the route. The path length of the planned path. and The weight values ​​corresponding to the time required to plan the path and the weight values ​​corresponding to the path length of the planned path are respectively assigned.

5. The twin factory AGV path planning method as described in claim 1, characterized in that, The acquisition of the cost function specifically includes: ； in, Let cost function be The actual cumulative cost from the starting point to n. Based on the cost of movement and =10, For the penalty function; in, ； in, The Manhattan distance from the starting point to the destination. As a penalty factor, For the intensity of punishment and .

6. The twin factory AGV path planning method as described in claim 1, characterized in that, YOLOv8n is used to detect and obtain the QR code of the automated guided vehicle (AGV) and upload it to the cloud server to synchronize the location information of the AGV.

7. A twin factory AGV path planning system, characterized in that, include: The building module is used to acquire map data of the factory to be planned and build a 3D twin factory model including automated guided vehicles (AGVs). The path planning module uses the shortest path planning time and the minimum planned path length as the objective function. Based on map data and the objective function, it employs the PSO-A* algorithm for path optimization to determine the globally optimal path planning result for the Automated Guided Vehicle (AGV) in the twin factory 3D model. The PSO-A* algorithm iteratively updates the weights of the heuristic function in the evaluation function of the improved A* algorithm by introducing a Particle Swarm Optimization (PSO) algorithm. This includes: initializing particles, assigning a weight to the heuristic function in the evaluation function for each particle; determining the initial fitness value of each particle based on the objective function; iteratively updating the position and velocity of each particle in the swarm using the PSO algorithm; determining the fitness value of each particle after the iterative update; using the weight value corresponding to the position of the particle with the optimal fitness value as the new weight of the heuristic function in the evaluation function of the improved A* algorithm; and outputting the corresponding path planning result. The improved A* algorithm optimizes the penalty function used to adjust the AGV's direction change in the cost function of the original A* algorithm based on a dynamic penalty mechanism. The velocity represents the direction and distance the particle moves in the next iteration.

8. A computer-readable storage medium, characterized in that, The storage medium stores a computer program, which, when executed by a processor, implements the twin factory AGV path planning method according to any one of claims 1-6.

9. A computer device, characterized in that, It includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the twin factory AGV path planning method according to any one of claims 1-6.

Images