Intelligent path planning method based on Unity3D and warehouse management system

Abstract

The invention relates to an intelligent path planning method based on Unity3D, which simulates a working environment by means of Unity3D, converts the working environment into a two-dimensional grid map, and guides and controls an AGV in combination with a corresponding path algorithm to ensure the stability and safety of the AGV delivery process. In addition, the invention also provides a warehouse management system which uses the above intelligent path planning method to realize planning of AGV distribution routes in a complex warehouse environment, optimize resource allocation in warehouse logistics links, reduce accidental loss in a material transportation process, improve safety and accuracy of logistics services, and improve the efficiency of the warehouse management system. Therefore, the management cost is reduced and the operation efficiency of the whole production system is continuously improved.

Description

Technical Field

[0001] The present invention relates to an intelligent path planning and management system for warehousing in the field of intelligent manufacturing, and specifically to an intelligent path planning method and a warehousing management system based on Unity3D. Background Art

[0002] Smart manufacturing relies on an efficient logistics system, and warehousing is a key link in the logistics chain. With the rapid popularization of informatization and the continuous expansion of the scale of the logistics industry, the traditional warehousing management model can no longer meet the development needs of the new generation of production and manufacturing. At present, there are many logistics tasks in the warehouse, the equipment operation logic is complex, the number of products is huge, and the status information of equipment, materials, environment, etc. changes with the progress of warehousing operations, making it difficult for warehouse operators to monitor and manage in real time. Especially when the stored materials and related in and out of the warehouse operations are more complicated, the information conveyed through the data table is relatively fragmented, and the material distribution path is subject to the constraints of actual conditions. It is often selected based on the personal experience and preferences of the relevant distribution personnel, lacking planning, resulting in material shortages, distribution delays, order retention and other phenomena.

[0003] Unity3D has a widely supported technology platform and hardware ecosystem. With its powerful graphics rendering capabilities and physics engine, it can build a virtual scene of physical warehouse entities and digitally simulate the entire process of warehouse operations. Path planning is one of the key technologies in the automation system. It can perform optimal trajectory planning based on a given environmental map, help mobile robots avoid obstacles, and optimize the system's resource allocation to ensure its efficiency and stability.

[0004] The main difficulty at present is that the path planning technology needs to be effectively integrated with the Unity3D physics engine and simulation environment to ensure the accuracy and stability of the distribution path. In addition, considering the dynamic obstacles in the warehousing environment and real-time environmental changes, real-time adjustment and optimization of the path need to be achieved. Summary of the invention

[0005] The present invention is an intelligent path planning method based on Unity3D, which simulates the working environment with the help of Unity3D and converts the working environment into a two-dimensional grid map, and guides and controls the AGV in combination with the corresponding path algorithm to ensure the stability and safety of the AGV distribution process; in addition, the present invention also provides a warehouse management system, which uses the above-mentioned intelligent path planning method to realize the planning of AGV distribution routes in complex warehousing environments, optimize resource allocation in warehousing logistics links, reduce accidental losses in the process of material transportation, and improve the safety and accuracy of logistics services, thereby reducing management costs and continuously improving the operating efficiency of the entire production system.

[0006] The present invention adopts the following technical solutions to achieve the above purpose: The intelligent path planning method based on Unity3D includes a Unity environment construction module, a map processing module, a path planning algorithm module, an AGV control module, and a data information processing module; the Unity environment construction module performs digital scene simulation on the working environment, the map processing module converts the digital simulation scene into a two-dimensional grid map, the path planning algorithm module calculates and plans the path according to the above two-dimensional grid map, the AGV control module guides the AGV according to the calculated and planned path, and the data processing module analyzes and processes the acquired data information and effectively transmits it to other functional modules.

[0007] The operation mode of the Unity environment construction module is to use modeling software and visualization tools to digitally present the layout of the actual working environment, including digital modeling of key physical entity elements in the working environment and digital simulation of static scenes such as the division of various areas of the working environment and the placement of equipment; at the same time, Unity3D is used to realize the digital expression of the workflow, including dynamic simulation of each work link, providing a full-dimensional, multi-dimensional working environment and process operation view.

[0008] The map processing module converts the virtual working environment layout in Unity3D into a two-dimensional grid map, providing basic map data support for the path planning module; identifies and marks obstacle elements in the map environment, including their specific location, shape, size and other information, so that the path planning module can perform obstacle avoidance processing; the map processing module is also equipped with an integrated sensor, which integrates sensor data to identify and record dynamic changes in the working environment, and updates the passable attributes of nodes in the map to ensure the accuracy of the map data.

[0009] The path planning algorithm module selects the JPS (Jump Point Search) algorithm as the core algorithm of the path planning module, and quickly finds the path with the lowest cost in the two-dimensional grid map to reach the target location through heuristic jump point search, thereby improving the distribution efficiency of the AGV; utilizes the structural characteristics of the priority queue to optimize the efficiency of processing nodes when the JPS algorithm finds the path, ensuring that the node with the lowest cost function value is always available for immediate access, thereby improving the response speed of the entire path planning algorithm; combines the map data provided by the map processing module and information such as dynamic obstacle position changes, to help the path planning module integrate dynamic obstacle avoidance and real-time update functions to cope with dynamic changes in the working environment, and ensure the stability and safety of the AGV distribution process.

[0010] The AGV control module receives the path result generated by the path planning algorithm module, controls the AGV to travel along the planned path in the virtual working environment of Unity3D, ensures that the AGV accurately reaches the target position, and effectively avoids collisions and conflicts; at the same time, a fault detection and feedback mechanism is established, and when an abnormal situation occurs, the emergency braking of the AGV is activated and an alarm is issued, and the fault information is recorded and transmitted to the data management module.

[0011] The data information processing module analyzes and processes the acquired data information and effectively transmits it to other functional modules, ensuring the collaborative work between various modules through data interaction and sharing; and provides data storage and logging functions, supports system persistence and data recovery, and records AGV operation logs, error information and system events to ensure the accuracy and stability of the path planning method.

[0012] The present invention also provides a warehouse management system, which uses the above-mentioned intelligent path planning method based on Unity3D, wherein the working environment targeted by each module is a warehouse operation environment.

[0013] The intelligent path planning method based on Unity3D includes the following functional modules and operation prevention: 1. Unity environment building module 1.1 Use modeling software and visualization tools to digitally present the layout of actual physical warehouses, including digital modeling of key physical elements in warehouses and digital simulation of static scenes such as warehouse area division and equipment placement; 1.2 Use Unity3D to realize the digital expression of warehouse operation process, including dynamic simulation of key links such as warehouse management, transportation task execution, and stacker operation, providing a full-dimensional and multi-dimensional view of warehouse operation; 2. Map processing module 2.1 Convert the virtual warehouse environment layout in Unity3D into a two-dimensional grid map to provide basic map data support for the path planning module; 2.2 Identify and mark obstacle elements in the map environment, including their specific location, shape, size and other information, so that the path planning module can perform obstacle avoidance processing; 2.3 Integrate sensor data to identify and record dynamic changes in the warehouse environment and update the traversable attributes of nodes in the map to ensure the accuracy of map data; 3. Path planning algorithm module 3.1 Select JPS (Jump Point Search) algorithm as the core algorithm of the path planning module. Through heuristic jump point search, it can quickly find the path with the lowest cost in the two-dimensional grid map to reach the target location, improve the distribution efficiency of AGV, and optimize the allocation of warehouse resources; 3.2 Utilize the structural characteristics of the priority queue to optimize the efficiency of processing nodes when the JPS algorithm is finding a path, ensuring that the node with the lowest cost function value is always available for immediate access, thereby improving the response speed of the entire path planning algorithm; 3.3 Combined with the map data provided by the map management module and information such as dynamic obstacle position changes, the path planning module is helped to integrate dynamic obstacle avoidance and real-time update functions to cope with dynamic changes in the warehouse environment and ensure the stability and safety of the AGV distribution process; 4. AGV control module 4.1 Receive the path results generated by the path planning module, control the AGV to drive along the planned path in the virtual warehouse environment of Unity3D, ensure that the AGV arrives at the target location accurately, and effectively avoid collisions and conflicts; 4.2 Establish a fault detection and feedback mechanism to start the emergency brake of the AGV and sound an alarm when an abnormal situation occurs, and record the fault information and pass it to the data management module; 5. Data information processing module 5.1 Analyze and process the acquired data information and effectively transmit it to other functional modules, ensuring the collaborative work between various modules through data interaction and sharing; 5.2 Provide data storage and logging functions, support system persistence and data recovery, record AGV operation logs, error information and system events to ensure the accuracy and stability of the path planning method.

[0014] Furthermore, the dynamic changes in the warehouse environment can be detected and updated in the following way: by adding a virtual sensor component to the automated guided vehicle (AGV) in the Unity warehouse environment, the built-in Collider component and Raycast function of Unity can be used to detect the position and movement of the AGV, and scripts can be written to obtain and update the position data of the AGV in real time; Collider and Rigidbody components can be added to the dynamic obstacle objects in the warehouse environment, and Unity's event system (such as OnTriggerEnter and OnTriggerExit) can be used to detect the position changes of dynamic obstacles, and the corresponding processing logic and position update can be triggered when the obstacle enters or leaves a specific area.

[0015] Furthermore, the path planning algorithm model is as follows: 1. After initializing the grid map, create an openlist and a closedlist. The openlist is used to store the jump points that need to be expanded each time, and the closelist is used to store the jump points that have been expanded. First, add the information of the starting node to the openlist and set its exploration priority to the highest value. 2. The data structure of the open list is designed using a priority queue and implemented through a minimum heap. When a new node is added to the open list, the node is first placed at the end of the heap, and then a float operation is performed to maintain the structural properties of the heap, so that the node with the lowest cost is at the position with the highest exploration priority, ensuring that the node with the lowest cost function value in the open list can be quickly accessed; 3. If the open list is not empty, calculate the traveled cost g(n), estimated cost h(n), and total cost f(n) of each node, and select the node with the highest priority from the open list, that is, the node with the lowest cost function value, as the current node; The traveled cost g(n) mentioned above represents the actual cost from the starting point to the current node; the estimated cost h(n) represents the estimated minimum cost from the current node to the target node, which can be calculated by different heuristic functions, such as Manhattan distance, Euclidean distance and Chebyshev distance. Here, Euclidean distance is selected, and its calculation formula is: The cost function value mentioned above, that is, the total cost f(n), is calculated as follows: 4. If the current node is the starting point, explore its subsequent jump points in the eight directions of straight lines and diagonal lines respectively; if the current node is not the starting point or the target point, explore its subsequent jump points along the vector direction of its parent node and the current node and the direction of the orthogonal component of the vector; when an obstacle, map boundary, or jump point is found in a certain direction, the search in that direction stops and the jump point is added to the open list. After the jump point search of the current node is completed, delete the current node from the open list and add it to the closed list; 5. If a jump point is already included in the open list, compare whether the g(n) of the jump point is less than the g(n) in the original open list. If so, update its parent node to the current node and recalculate its cost values ​​h(n), g(n) and f(n); 6. Iteratively select the jump point with the smallest cost in the open list as the next node to be expanded, and repeat step 4 until the target point already exists in the closed list, or the search for the target point fails (that is, the open list is empty), then the current algorithm ends; 7. Starting from the target point, traverse the parent node chain in reverse order, build a complete path from the starting point to the target point, and obtain the optimal path result.

[0016] The present invention has the following significant advantages: 1. The present invention realizes the effective integration of the warehouse environment and the path planning algorithm in Unity3D. By combining the intuitive visual editor of Unity3D and the optimization capability of the path planning algorithm, the whole process of warehouse operation can be effectively monitored and the distribution path of the warehouse AGV can be optimized, thus further promoting the development of intelligent management of warehouse.

[0017] 2. The present invention designs a reasonable path planning strategy to cope with the dynamic changes in the warehousing environment. The AGV path planning algorithm is used to improve the rationality and stability of the transportation route, reduce the occurrence of conflicts during material transportation, improve logistics operation efficiency, alleviate warehousing pressure during business peak periods, and reduce transportation costs. It provides a reference solution for promoting the application of intelligent mobile robots in different environments. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 This is a schematic diagram of the architecture of the intelligent path planning method based on Unity3D; Figure 2 This is a model diagram of the path planning algorithm of the present invention. DETAILED DESCRIPTION

[0019] The present invention is further described below in conjunction with the accompanying drawings and examples. It should be clear that the following examples are only one or more manifestations of the present invention and are not a complete limitation of the invention described in this application. Any improvements known to those skilled in the art made on the basis of the invention or technical solutions described in this application should fall within the scope of protection claimed in this application.

[0020] In order to more clearly describe the objectives, technical solutions and advantages in this application, further detailed description will be provided in combination with the accompanying drawings and actual examples.

[0021] In one embodiment, a warehouse management system is provided, which adopts an intelligent path planning method based on Unity3D, combined with Figure 1 ,The specific steps of the intelligent planning method are as follows: Step 1: Build a virtual storage environment 1.1 Use the modeling software Blender to build three-dimensional models of key elements in physical warehousing, such as stackers, automatic guided vehicles (AGVs), goods, shelves, etc.; 1.2 Import the created 3D model into Unity3D, and use the visual editor of Unity3D to build a virtual warehouse scene, and perform visual simulation of the layout of equipment placement, shelf layout, cargo storage, loading and unloading areas, picking areas, storage areas, etc. in the warehouse; 1.3 In the virtual warehousing scene, with the help of Unity3D's powerful physics engine and program scripts written in C# language, the model is driven to complete dynamic simulation of scenes such as cargo picking, transportation task execution, and stacker operation.

[0022] Step 2: Dynamic obstacle position detection and update 2.1 Add virtual sensor components to the automated guided vehicle (AGV) in the Unity warehouse environment. You can use Unity's built-in Collider component and Raycast function to detect the position and movement of the AGV, and write scripts to obtain and update the AGV's position data in real time; 2.2 Add Collider and Rigidbody components to dynamic obstacle objects in the warehouse environment, and use Unity's event system (such as OnTriggerEnter and OnTriggerExit) to detect the position changes of dynamic obstacles, triggering corresponding processing logic and position updates when obstacles enter or leave specific areas.

[0023] Step 3: Map environment modeling 3.1 Convert the spatial information of the warehouse environment in Unity3D into a graphical data structure, generate a two-dimensional grid map based on the width, length and obstacle list of the warehouse scene, traverse each position in the scene, and generate corresponding grid nodes; 3.2 Initialize the attributes of each cell on the two-dimensional grid map and check whether it overlaps with the location coordinates of the obstacle, identify the location coordinates of each node on the map and whether it is passable; 3.3 With the help of virtual sensor components, the static obstacle positions and dynamic obstacle position changes in the storage environment are detected and marked, and the passability of the corresponding nodes in the grid map is updated in real time to ensure the accuracy of the path planning results.

[0024] Step 4: Build a path planning algorithm model 4.1 The specific process of the algorithm is as follows Figure 2As shown, after initializing the grid map, create an open list (openlist) and a closed list (closelist). The open list is used to store the jump points that need to be expanded each time, and the close list is used to store the jump points that have been expanded. First, add the information of the starting node to the open list (openlist) and set its exploration priority to the highest value. 4.2 The data structure of the open list is designed using a priority queue and implemented through a minimum heap. When a new node is added to the open list, the node is first placed at the end of the heap, and then a float operation is performed to maintain the structural properties of the heap, so that the node with the lowest cost is at the position with the highest exploration priority, ensuring that the node with the lowest cost function value in the open list can be quickly accessed; 4.3 If the open list is not empty, calculate the traveled cost g(n), estimated cost h(n), and total cost f(n) of each node, and select the node with the highest priority from the open list, that is, the node with the lowest cost function value, as the current node; The traveled cost g(n) mentioned above represents the actual cost from the starting point to the current node; the estimated cost h(n) represents the estimated minimum cost from the current node to the target node, which can be calculated by different heuristic functions, such as Manhattan distance, Euclidean distance and Chebyshev distance. Here, Euclidean distance is selected, and its calculation formula is: The cost function value mentioned above, that is, the total cost f(n), is calculated as follows: 4.4 If the current node is the starting point, explore its subsequent jump points in the eight directions of straight lines and diagonal lines respectively; if the current node is not the starting point or the target point, explore its subsequent jump points along the vector direction of its parent node and the current node and the direction of the vector orthogonal component; when an obstacle, map boundary, or jump point is found in a certain direction, the search in that direction stops and the jump point is added to the open list. After the jump point search of the current node is completed, delete the current node from the open list and add it to the closed list; 4.5 If a certain jump point is already included in the open list, compare whether the g(n) of the jump point is less than the g(n) in the original open list. If so, update its parent node to the current node and recalculate its cost values ​​h(n), g(n) and f(n); 4.6 Iteratively select the jump point with the smallest cost in the open list as the next node to be expanded, and repeat step 4.4 until the target point already exists in the closed list, or the search for the target point fails (that is, the open list is empty), then end the current algorithm; 4.7 Starting from the target point, traverse the parent node chain in reverse order, construct a complete path from the starting point to the target point, and obtain the optimal path result.

[0025] Step 5: AGV navigation control 5.1 Receive and parse the optimal path result from the path planning module, extract the path point data and encapsulate it into a JSON format data structure, and use each array item to represent the specific AGV control command, which contains the coordinates of each path point, steering angle and other information; 5.2 The integrated sensor component obtains the AGV position coordinates and surrounding environment information in real time to identify potential faults or abnormal behaviors, such as collisions, track deviations, etc., and records the fault information, while immediately stopping the movement of the AGV to ensure the safety of equipment and goods.

[0026] Step 6: Data transmission and log management 6.1 Use Unity's event and delegation mechanism to transmit the path planning results. The AGV control module first subscribes to the corresponding event. When the path planning algorithm obtains the result, the event is triggered. The AGV control module can update its navigation system in real time.

[0027] 6.2 Use Unity's built-in logging system and use methods such as Debug.Log() in the script to record operation logs, error information, etc., and serialize key data objects into JSON strings through JsonUtility, and then store them in the file system to achieve persistent storage of key data.

Claims

1. An intelligent path planning method based on Unity3D, characterized by: It includes a Unity environment construction module, a map processing module, a path planning algorithm module, an AGV control module, and a data information processing module; the Unity environment construction module performs digital scene simulation on the working environment, the map processing module converts the digital simulation scene into a two-dimensional grid map, the path planning algorithm module calculates and plans the path according to the above two-dimensional grid map, the AGV control module guides the AGV according to the calculated and planned path, and the data processing module analyzes and processes the acquired data information and effectively transmits it to other functional modules.

2. The intelligent path planning method based on Unity3D according to claim 1, characterized in that: The operation mode of the Unity environment construction module is to use modeling software and visualization tools to digitally present the layout of the actual working environment, including digital modeling of key physical entity elements in the working environment and digital simulation of static scenes such as the division of various areas of the working environment and the placement of equipment; at the same time, Unity3D is used to realize the digital expression of the workflow, including dynamic simulation of each work link, providing a full-dimensional, multi-dimensional working environment and process operation view.

3. The intelligent path planning method based on Unity3D according to claim 1, characterized in that: The map processing module converts the virtual working environment layout in Unity3D into a two-dimensional grid map, providing basic map data support for the path planning module; identifies and marks obstacle elements in the map environment, including their specific location, shape, size and other information, so that the path planning module can perform obstacle avoidance processing; the map processing module is also equipped with an integrated sensor, which integrates sensor data to identify and record dynamic changes in the working environment, and updates the passable attributes of nodes in the map to ensure the accuracy of the map data.

4. The intelligent path planning method based on Unity3D according to claim 1, characterized in that: The path planning algorithm module selects the JPS (Jump Point Search) algorithm as the core algorithm of the path planning module, and quickly finds the path with the lowest cost in the two-dimensional grid map to reach the target location through heuristic jump point search, thereby improving the distribution efficiency of the AGV; utilizes the structural characteristics of the priority queue to optimize the efficiency of processing nodes when the JPS algorithm finds the path, ensuring that the node with the lowest cost function value is always available for immediate access, thereby improving the response speed of the entire path planning algorithm; combines the map data provided by the map processing module and information such as dynamic obstacle position changes, to help the path planning module integrate dynamic obstacle avoidance and real-time update functions to cope with dynamic changes in the working environment, and ensure the stability and safety of the AGV distribution process.

5. The intelligent path planning method based on Unity3D according to claim 1, characterized in that: The AGV control module receives the path result generated by the path planning algorithm module, controls the AGV to travel along the planned path in the virtual working environment of Unity3D, ensures that the AGV accurately reaches the target position, and effectively avoids collisions and conflicts; at the same time, a fault detection and feedback mechanism is established, and when an abnormal situation occurs, the emergency braking of the AGV is activated and an alarm is issued, and the fault information is recorded and transmitted to the data management module.

6. The intelligent path planning method based on Unity3D according to claim 1, characterized in that: The data information processing module analyzes and processes the acquired data information and effectively transmits it to other functional modules, ensuring the collaborative work between the modules through data interaction and sharing; It also provides data storage and logging functions, supports system persistence and data recovery, and records AGV operation logs, error messages, and system events to ensure the accuracy and stability of the path planning method.

7. A warehouse management system, which uses the Unity3D-based intelligent path planning method described in any one of claims 1-6, wherein the working environment targeted by each module is a warehouse operation environment.

8. The intelligent path planning method based on Unity3D according to claim 7, characterized in that: The intelligent path planning method based on Unity3D includes the following functional modules and operation prevention:

1. Unity environment building module 1.1 Use modeling software and visualization tools to digitally present the layout of actual physical warehouses, including digital modeling of key physical elements in warehouses and digital simulation of static scenes such as warehouse area division and equipment placement; 1.2 Use Unity3D to realize the digital expression of warehouse operation process, including dynamic simulation of key links such as warehouse management, transportation task execution, and stacker operation, providing a full-dimensional and multi-dimensional view of warehouse operation; 2. Map processing module 2.1 Convert the virtual warehouse environment layout in Unity3D into a two-dimensional grid map to provide basic map data support for the path planning module; 2.2 Identify and mark obstacle elements in the map environment, including their specific location, shape, size and other information, so that the path planning module can perform obstacle avoidance processing; 2.3 Integrate sensor data to identify and record dynamic changes in the warehouse environment and update the traversable attributes of nodes in the map to ensure the accuracy of map data; 3. Path planning algorithm module 3.1 Select JPS (Jump Point Search) algorithm as the core algorithm of the path planning module. Through heuristic jump point search, it can quickly find the path with the lowest cost in the two-dimensional grid map to reach the target location, improve the distribution efficiency of AGV, and optimize the allocation of warehouse resources; 3.2 Utilize the structural characteristics of the priority queue to optimize the efficiency of processing nodes when the JPS algorithm is finding a path, ensuring that the node with the lowest cost function value is always available for immediate access, thereby improving the response speed of the entire path planning algorithm; 3.3 Combined with the map data provided by the map management module and information such as dynamic obstacle position changes, the path planning module is helped to integrate dynamic obstacle avoidance and real-time update functions to cope with dynamic changes in the warehouse environment and ensure the stability and safety of the AGV distribution process; 4. AGV control module 4.1 Receive the path results generated by the path planning module, control the AGV to drive along the planned path in the virtual warehouse environment of Unity3D, ensure that the AGV arrives at the target location accurately, and effectively avoid collisions and conflicts; 4.2 Establish a fault detection and feedback mechanism to start the emergency brake of the AGV and sound an alarm when an abnormal situation occurs, and record the fault information and pass it to the data management module; 5. Data information processing module 5.1 Analyze and process the acquired data information and effectively transmit it to other functional modules, ensuring the collaborative work between various modules through data interaction and sharing; 5.2 Provide data storage and logging functions, support system persistence and data recovery, record AGV operation logs, error information and system events to ensure the accuracy and stability of the path planning method.

9. The intelligent path planning method based on Unity3D according to claim 8, characterized in that: The method for detecting and updating dynamic changes in the warehouse environment is as follows: add virtual sensor components to the automated guided vehicle AGV in the Unity warehouse environment, use Unity's built-in Collider components and Raycast functions to detect the position and movement of the AGV, and write scripts to obtain and update the AGV's position data in real time; add Collider and Rigidbody components to dynamic obstacle objects in the warehouse environment, and use Unity's event system (such as OnTriggerEnter and OnTriggerExit) to detect position changes of dynamic obstacles, and trigger corresponding processing logic and position updates when obstacles enter or leave specific areas.

10. The intelligent path planning method based on Unity3D according to claim 8, characterized in that: The path planning algorithm model is as follows:

1. After initializing the grid map, create an openlist and a closedlist. The openlist is used to store the jump points that need to be expanded each time, and the closelist is used to store the jump points that have been expanded. First, add the information of the starting node to the openlist and set its exploration priority to the highest value.

2. The data structure of the open list is designed using a priority queue and implemented through a minimum heap. When a new node is added to the open list, the node is first placed at the end of the heap, and then a float operation is performed to maintain the structural properties of the heap, so that the node with the lowest cost is at the position with the highest exploration priority, ensuring that the node with the lowest cost function value in the open list can be quickly accessed; 3. If the open list is not empty, calculate the traveled cost g(n), estimated cost h(n), and total cost f(n) of each node, and select the node with the highest priority from the open list, that is, the node with the lowest cost function value, as the current node; The traveled cost g(n) mentioned above represents the actual cost from the starting point to the current node; The estimated cost h(n) represents the estimated minimum cost from the current node to the target node, which can be calculated by different heuristic functions, such as Manhattan distance, Euclidean distance, and Chebyshev distance. Here, Euclidean distance is selected, and its calculation formula is: The cost function value mentioned above, that is, the total cost f(n), is calculated as follows:

4. If the current node is the starting point, explore its successor jump points in the eight directions of straight lines and oblique lines respectively; if the current node is not the starting point or the target point, explore its successor jump points along the vector direction of its parent node and the current node and the direction of the orthogonal component of the vector; When an obstacle, map boundary, or jump point is encountered in a certain direction, the search in that direction stops and the jump point is added to the open list. After the jump point search of the current node is completed, the current node is deleted from the open list and added to the closed list; 5. If a certain jump point is already included in the open list, compare whether the g(n) of the jump point is less than the g(n) in the original open list; If this happens, update its parent node to the current node and recalculate its cost values ​​h(n), g(n) and f(n); 6. Iteratively select the jump point with the smallest cost in the open list as the next node to be expanded, and repeat step 4 until the target point already exists in the closed list, or the search for the target point fails (that is, the open list is empty), then the current algorithm ends; 7. Starting from the target point, traverse the parent node chain in reverse order, build a complete path from the starting point to the target point, and obtain the optimal path result.

Images