ROS2-based traffic control and scheduling method of distributed AGV system

Abstract

The invention provides a distributed AGV system traffic control and scheduling method realized based on ROS2, and the method comprises the steps: building AGV vehicle nodes, building a traffic control module, developing a scheduling module, building a communication network, integrating a sensor data processing module, providing a central control board / man-machine interaction interface, and building a database / log module. According to the method, efficient data interaction and collaborative operation between AGVs are realized by using a distributed communication framework of ROS2, and by constructing a traffic control module and an optimized scheduling algorithm, the position and state of each AGV can be monitored in real time, the running path of each AGV can be dynamically adjusted, and traffic congestion and collision events are avoided. Compared with a traditional centralized control system, the system has the advantages that the expandability and fault-tolerant capability of the system are improved, the system is suitable for a complex and changeable industrial logistics environment, and the efficiency and reliability of logistics transportation are remarkably improved.

Description

Technical Field

[0001] The present invention relates to the field of intelligent transportation and automatic control technology, and in particular to a traffic control and dispatching method of a distributed AGV system based on ROS2. Background Art

[0002] With the deepening of intelligent manufacturing, automated guided vehicles (AGVs) have been widely used in logistics, warehouse management, and manufacturing. The AGV system realizes the automatic handling and distribution of materials, improves production efficiency, and reduces labor costs. However, traditional AGV systems usually adopt a centralized control method, and the central controller uniformly manages the path planning and task scheduling of all AGVs. This centralized architecture has problems such as limited scalability, single point failure risk, and slow response speed. When the number of AGVs increases, the load of the central controller increases significantly and the system performance decreases; once the central controller fails, it may cause the entire system to paralyze; in a complex production environment, the centralized system responds slowly to environmental changes and cannot meet real-time requirements.

[0003] In order to solve the above problems, distributed AGV systems have gradually become a hot topic in research and application. In a distributed system, each AGV has the ability to make autonomous decisions and collaborate, which reduces the burden on the central controller and improves the flexibility and fault tolerance of the system. However, the implementation of a distributed AGV system also faces many challenges, including communication and collaboration problems, traffic control complexity, and scheduling algorithm optimization. In a complex industrial environment, it is very difficult to implement an efficient and reliable communication mechanism to ensure information sharing and collaborative operations among multiple AGVs; how to effectively control traffic and avoid path conflicts and traffic congestion is an urgent problem to be solved; existing scheduling algorithms have deficiencies in real-time and computational complexity, and are difficult to adapt to dynamic production tasks and environments.

[0004] In recent years, ROS2 (Robot Operating System 2), as a new generation of robot operating system, has attracted widespread attention due to its distributed architecture, real-time support and efficient communication mechanism. ROS2 uses DDS (Data Distribution Service) as the communication middleware, providing a reliable publish / subscribe model suitable for high-concurrency and real-time application scenarios. Building a distributed AGV system based on ROS2 has the advantages of high scalability, high reliability and real-time communication. However, the traffic control and scheduling methods of the distributed AGV system based on ROS2 are still in the exploratory stage. There are problems such as lack of a unified traffic control mechanism, imperfect scheduling strategy and low system integration, which cannot give full play to the overall performance of the system.

[0005] In view of the above background, there is an urgent need for a traffic control and scheduling method for a distributed AGV system based on ROS2. Summary of the invention

[0006] In order to solve various problems commonly existing in the prior art, the purpose of the present invention is to provide a traffic control and scheduling method for a distributed AGV system based on ROS2. The invention makes full use of the distributed communication framework and real-time support of ROS2 to build an efficient and reliable traffic control mechanism and intelligent scheduling algorithm, realize the collaborative operation of multiple AGV systems, improve the operating efficiency and safety of the system, and meet the needs of modern industry for intelligent logistics and automated production.

[0007] The present invention achieves the above-mentioned purpose through the following technical solutions: A traffic control and dispatching method for a distributed AGV system based on ROS2 includes the following steps: Establish AGV vehicle nodes: configure each AGV as an independent ROS2 node, and integrate the synchronous positioning and mapping algorithm to achieve autonomous positioning and mapping; install a sensor unit on each AGV to obtain surrounding environment information; each AGV regularly publishes its own status information through the ROS2 communication mechanism; Constructing traffic control module: Establishing a traffic control node on the central server as a global traffic monitoring center; this node subscribes to the status information of all AGV vehicle nodes and implements conflict detection algorithms to predict potential traffic conflicts; when potential conflicts are detected, control instructions are sent to relevant AGV vehicle nodes; at the same time, priority strategies are formulated based on priority factors to determine the traffic priority of each AGV vehicle node; Develop a scheduling module: receive and manage task requests from different sources; obtain the real-time status of all AGV vehicle nodes and evaluate their availability and applicability; use the path planning algorithm to calculate the shortest driving path / optimal driving path for the AGV vehicle nodes; assign AGVs to perform tasks based on the status of the AGV vehicle nodes, task requirements, and path planning results, and send corresponding task instructions; Establish a communication network: Use the DDS communication mechanism of ROS2 to realize data exchange and communication between AGV vehicle nodes, traffic control modules and scheduling modules; Integrated sensor data processing module: collects and processes data from the sensor units of AGV vehicle nodes, integrates, analyzes and processes the collected data, and provides environmental perception and decision support for the traffic control module and the dispatching module; Provide central control console / human-computer interaction interface: Design and implement a central control console or human-computer interaction interface to display the AGV's operating status, task progress, and system alarm information; Establish database / log module: Establish a database system to store historical task data, AGV operation logs and system events, and provide data query, analysis and export functions.

[0008] According to a traffic control and scheduling method for a distributed AGV system based on ROS2 provided by the present invention, when implementing a conflict detection algorithm, by continuously subscribing to the ROS2 topics published by all AGV vehicle nodes, the location information, operating status, current task endpoint and travel direction of each AGV vehicle node are received and updated in real time; Based on the current position, speed, direction and task destination of each AGV vehicle node, a path planning algorithm is used to predict its path trajectory in the future. At the same time, a conflict detection algorithm is implemented to compare the spatial position relationship of each AGV predicted path to identify possible path intersections or approaches, that is, potential collision risks. When the conflict detection algorithm predicts that the paths of any two or more AGV vehicle nodes will intersect or approach to within a preset safety distance at a certain time point or spatial area in the future, the conflict warning mechanism is immediately triggered; the warning information includes the type of conflict, the expected time and location, the AGV numbers involved, and the recommended avoidance measures; According to the preset priority strategy and rules, determine the AGV vehicle nodes that need to adjust the path; use the dynamic path replanning algorithm to calculate new feasible paths for the affected AGV vehicle nodes; The calculated new path is sent to the relevant AGV vehicle node in the form of ROS2 service or action message, instructing it to drive along the new path; at the same time, the execution of the adjustment instruction by the AGV vehicle node is continuously monitored, including the path tracking accuracy, speed control and the implementation effect of the avoidance measures.

[0009] According to a traffic control and dispatching method for a distributed AGV system implemented based on ROS2 provided by the present invention, when a dispatching module is called, the module is used to manage a dynamically updated task queue, which contains all tasks to be executed and related information; the task queue is sorted in real time according to priority; For each task to be executed, the scheduling module selects an AGV vehicle node with corresponding capabilities and resources from the available AGV list for matching according to the specific requirements of the task; an intelligent matching algorithm is used to comprehensively consider the current position, status, remaining power, and task history factors of the AGV vehicle node to achieve the optimal matching between the task and the AGV vehicle node; For AGV vehicle nodes that have been assigned tasks, the scheduling module uses the A* algorithm to calculate the optimal path from the current position to the task end point; at the same time, the path planning results are combined with the dynamic characteristics of the AGV to smooth the path and generate an executable driving path; During the AGV's task execution, the scheduling module continuously monitors the AGV's status changes and environmental dynamics; when an abnormal situation or environmental change is detected, the dynamic scheduling mechanism is immediately triggered to re-evaluate the task queue, AGV capabilities, and path planning results, and make necessary adjustments and optimizations.

[0010] According to a traffic control and dispatching method of a distributed AGV system based on ROS2 provided by the present invention, the following steps are also performed: Establish a task execution feedback mechanism to collect feedback information from AGV vehicle nodes in real time during the execution of tasks; conduct in-depth analysis and processing of the collected feedback information, and use data analysis algorithms and machine learning models to identify bottlenecks, inefficient links, and potential risk points in the task execution process; based on the analysis results of the feedback data, the scheduling module dynamically adjusts the task allocation strategy, path planning algorithm, and AGV scheduling rules for continuous optimization and adjustment; at the same time, each strategy adjustment based on feedback information is recorded, including performance comparison before and after the adjustment, specific measures for the adjustment, implementation time, and evaluation of the adjustment effect, to provide reference and basis for subsequent task scheduling and path planning.

[0011] According to a traffic control and dispatching method for a distributed AGV system based on ROS2 provided by the present invention, when establishing a communication network, a functional analysis is performed on each ROS2 node in the system, and the nodes are classified according to the tasks and roles they undertake; based on the functional classification, the nodes are modularly laid out, and nodes that perform similar or related tasks are placed in close positions in the network topology; Evaluate the communication requirements of each ROS2 node, including data transmission volume, real-time requirements, and reliability requirements; allocate network bandwidth resources based on the communication demand assessment results to ensure that key nodes and communication links have sufficient bandwidth; For node pairs that require high-frequency, large-volume communications, a point-to-point connection is used to reduce communication delays and packet loss rates; Utilizing the multicast communication function of ROS2, for scenarios where the same data needs to be broadcast to multiple nodes, a multicast connection method is used to implement a one-to-many data transmission mode, where multiple AGV vehicle nodes receive the same control instructions or data updates at the same time. The QoS configuration function of ROS2 is used to optimize the communication behavior of ROS2 nodes.

[0012] According to a traffic control and dispatching method of a distributed AGV system implemented based on ROS2 provided by the present invention, when using a sensor data processing module, data from various sensor units of each AGV vehicle node is collected in real time, and the collected data is transmitted to a central processing unit; Preprocess the collected raw data, including data cleaning, denoising, calibration, formatting, and time stamping; Fuse the data from the sensor units of different AGV vehicle nodes, and conduct in-depth analysis and processing on the fused data; apply machine learning or deep learning algorithms to mine and model the fused data to extract valuable information and patterns; Based on the analyzed and processed data, environmental perception information is generated, decision support suggestions are generated, and real-time and accurate environmental descriptions are provided for traffic control modules and scheduling modules.

[0013] According to a traffic control and dispatching method for a distributed AGV system based on ROS2 provided by the present invention, when establishing an AGV vehicle node, an independent ROS2 node is configured for each AGV and basic parameters of the ROS2 node are configured, and each AGV communicates and exchanges data as an independent entity in the network; Integrate the sensor unit on the AGV to obtain environmental perception data; Select and integrate the SLAM algorithm suitable for AGV and configure the SLAM algorithm parameters; The time synchronization mechanism of ROS2 is used to ensure that the data from different sensors are consistent in time, and the data from different sensors are fused through the data fusion algorithm; Start the SLAM algorithm and use sensor data to perform autonomous positioning and map construction of the AGV in real time; The constructed map is stored in the global map server of ROS2 for access and use by other nodes.

[0014] According to a traffic control and dispatching method of a distributed AGV system based on ROS2 provided by the present invention, the following steps are also performed: Continuously monitor nodes and monitor the operating status of all AGV vehicle nodes and server nodes in real time; Use the heartbeat mechanism or regular status reports to ensure that the status information of each node is updated in real time and fed back to the system monitoring center; When a fault is detected in an AGV vehicle node or server node, the fault detection mechanism is immediately triggered to quickly locate and diagnose the fault; Automatically execute failover strategies based on the fault type and severity, including switching to pre-configured backup nodes and reallocating tasks to other normally operating AGVs; After a failover, tasks are automatically reallocated and scheduled based on the current task status and AGV availability to ensure that all tasks are processed and completed in a timely manner.

[0015] According to a traffic control and scheduling method of a distributed AGV system implemented based on ROS2 provided by the present invention, when the A* algorithm is used to calculate the optimal path from the current position to the task end point, it specifically includes: Initialize path search: determine the current position of the AGV as the starting point of the path search, and the task end point as the target position; initialize an Open List to store the nodes to be evaluated, and a Closed List to store the nodes that have been evaluated; add the starting point node to the Open List, and set its cost to 0 and the parent node to be empty; Define node cost and heuristic function: For each node, define its cost as the actual path length from the starting point to the node; define a heuristic function to estimate the minimum cost from the current node to the target position; Node evaluation: Select the node with the smallest sum of cost and heuristic value from the open list as the current node for evaluation; remove the current node from the open list and add it to the closed list, indicating that the node has been evaluated; Expand the current node: Based on the position and environment information of the current node, generate all possible adjacent nodes, that is, the next position that can be moved to; for each adjacent node, calculate its cost as the cost of the current node plus the cost of moving from the current node to the adjacent node; if the adjacent node is in the closed list, ignore the node; otherwise, add it to the open list and set its parent node to the current node.

[0016] According to a traffic control and scheduling method for a distributed AGV system implemented based on ROS2 provided by the present invention, an open list is updated: the nodes in the open list are sorted to ensure that the node with the smallest sum of cost and heuristic value is always at the front of the list; the steps of node evaluation and selection and expansion of the current node are repeated until the target position is added to the open list or the open list is empty; Construct the optimal path: If the target location is added to the open list, start from the target location and trace back to the starting point through its parent node step by step to construct the optimal path from the current location to the end point of the task; if the open list is empty and the target location is not found, it means that the target location cannot be reached, and an empty path or error message should be returned at this time; Smoothing the constructed optimal path and removing unnecessary turning points or redundant nodes to simplify the path and improve driving efficiency; During the driving process of the AGV, the path is dynamically adjusted according to the real-time updated environmental information and the current state of the AGV, and the adjusted path is sent to the actuator of the AGV to control the AGV to drive according to the adjusted path.

[0017] It can be seen that compared with the prior art, the present invention has the following beneficial effects: 1. The present invention distributes control and decision-making tasks to multiple nodes, thereby reducing the burden on the central controller and enabling it to handle core tasks more efficiently. It also facilitates the expansion and reduction of the system, and flexibly adjusts the number of AGVs according to actual needs to meet application requirements of different scales.

[0018] 2. By implementing a fault-tolerant mechanism and data redundancy strategy, the present invention can continue to operate when a single node or component fails, thereby significantly improving the overall reliability and stability and reducing system downtime caused by single point failures.

[0019] 3. The AGV node of the present invention has the ability to make independent decisions, can respond quickly according to actual conditions and emergencies, and can meet real-time requirements. It is crucial for meeting application scenarios with high real-time requirements, such as dynamically changing warehouse environments.

[0020] 4. The present invention utilizes efficient traffic control strategies and intelligent scheduling algorithms to optimize the driving paths of AGVs, thereby reducing path conflicts and traffic congestion between AGVs, improving task execution efficiency, and ensuring smooth operation of the system.

[0021] 5. The present invention realizes reliable data exchange among AGV nodes, traffic control modules and dispatching modules through the efficient communication mechanism of ROS2, ensures the coordinated operation among various modules and improves the overall performance of the system.

[0022] 6. The central console interface and human-computer interaction interface are designed to be user-friendly, allowing operators to easily monitor and manage the system. This not only improves work efficiency, but also reduces the difficulty of operation, making the system easier to accept and use.

[0023] 7. The sensor data processing module enhances the system's environmental perception capabilities, enabling it to more accurately understand the surrounding environment and make corresponding adjustments. At the same time, the database / log module supports big data analysis, providing strong support for system optimization and decision-making, which helps to discover potential problems, optimize system performance and improve overall efficiency.

[0024] 8. The present invention uses the A* algorithm to calculate the optimal driving path of the AGV, avoids known obstacles and high traffic density areas, considers dynamic environmental factors, and optimizes the path in real time. The present invention establishes a dynamic scheduling and feedback mechanism to update task allocation and path planning in real time when the AGV status or environment changes. By monitoring the execution of tasks and collecting feedback information from AGVs, the scheduling strategy is adjusted in a timely manner to improve the adaptability of the system.

[0025] The present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 It is a schematic diagram of an embodiment of a traffic control and dispatching method of a distributed AGV system implemented based on ROS2 of the present invention.

[0027] Figure 2 It is a flow chart of an embodiment of a traffic control and dispatching method of a distributed AGV system implemented based on ROS2 of the present invention. DETAILED DESCRIPTION

[0028] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0029] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0030] A traffic control and dispatching method embodiment of a distributed AGV system based on ROS2 See also Figure 1 and Figure 2 This embodiment provides a traffic control and scheduling method for a distributed AGV system based on ROS2, comprising the following steps: Establish AGV vehicle nodes: configure each AGV as an independent ROS2 node, and integrate synchronous positioning and map building algorithms to achieve autonomous positioning and map building; install a sensor unit on each AGV to obtain surrounding environment information; each AGV regularly publishes its own status information through the ROS2 communication mechanism, including but not limited to position, speed, power and load status.

[0031] Build a traffic control module: Establish a traffic control node on the central server as a global traffic monitoring center; this node subscribes to the status information of all AGV vehicle nodes and implements a conflict detection algorithm to predict potential traffic conflicts; when a potential conflict is detected, it sends control instructions to the relevant AGV vehicle nodes, such as slowing down, stopping, or re-planning the path; at the same time, formulate a priority strategy based on priority factors to determine the traffic priority of each AGV vehicle node.

[0032] Develop a scheduling module: receive and manage task requests from different sources, each task request contains information such as the starting point, end point, empty / loaded type, and priority; obtain the real-time status of all AGV vehicle nodes and evaluate their availability and applicability; use the path planning algorithm to calculate the shortest driving path / optimal driving path for the AGV vehicle nodes, while considering avoiding obstacles and high traffic density areas; assign AGVs to perform tasks according to the status of the AGV vehicle nodes, task requirements, and path planning results, and send corresponding task instructions; Establish a communication network: Use the DDS communication mechanism of ROS2 to realize data exchange and communication between AGV vehicle nodes, traffic control modules and scheduling modules.

[0033] Integrated sensor data processing module: collects and processes data from the sensor units of AGV vehicle nodes, including but not limited to data from lidar, ultrasonic sensors, cameras, etc.; fuses, analyzes and processes the collected data to provide environmental perception and decision support for the traffic control module and scheduling module.

[0034] Provide a central console / human-computer interaction interface: Design and implement a central console or human-computer interaction interface to display important information such as the AGV's operating status, task progress, and system alarm information; allow operators to perform manual intervention operations such as system monitoring, task management, and AGV control through this interface.

[0035] Establish database / log module: Establish a database system to store historical task data, AGV operation logs and system events, and provide data query, analysis and export functions to provide data support for system optimization, fault diagnosis and performance evaluation.

[0036] In this embodiment, when implementing the conflict detection algorithm, by continuously subscribing to the ROS2 topics published by all AGV vehicle nodes, the location information, operating status, current task endpoint and travel direction of each AGV vehicle node are received and updated in real time; high-precision positioning technology and time synchronization mechanism are used to ensure the accuracy and consistency of the data.

[0037] Based on the current position, speed, direction and task endpoint of each AGV vehicle node, a path planning algorithm is used to predict its path trajectory in the future; at the same time, a conflict detection algorithm is implemented to compare the spatial position relationship of each AGV predicted path to identify possible path intersections or approaches, that is, potential collision risks.

[0038] When the conflict detection algorithm predicts that the path trajectories of any two or more AGV vehicle nodes will intersect or approach to within a preset safety distance at a certain time point or spatial area in the future, the conflict warning mechanism is immediately triggered; the warning information includes the type of conflict, the expected time and location of the conflict, the AGV number involved, and the recommended avoidance measures. Among them, the conflict warning mechanism of this embodiment includes: setting a safety distance threshold, when it is predicted that the path trajectories of any two AGVs will be less than the safety distance at a certain moment in the future, the conflict warning is triggered; according to the type and severity of the conflict, the warning is divided into different levels, such as minor, moderate and severe, and corresponds to different response strategies.

[0039] According to preset priority strategies and rules, such as task urgency, AGV load, path length or energy consumption, the AGV vehicle nodes that need to adjust the path are determined; a dynamic path replanning algorithm is used to calculate new feasible paths for the affected AGV vehicle nodes to avoid conflicts and ensure efficient execution of tasks; in the path adjustment process, global optimization objectives are considered, such as minimizing the total driving time, total energy consumption or total path length.

[0040] The calculated new path is sent to the relevant AGV vehicle node in the form of ROS2 service or action message, instructing it to drive along the new path; at the same time, the execution of the adjustment instructions by the AGV vehicle node is continuously monitored, including the path tracking accuracy, speed control and the implementation effect of avoidance measures, to ensure that conflicts are effectively avoided and the system runs safely and stably.

[0041] Specifically, the traffic control module of this embodiment establishes a traffic control node in the central server in advance, subscribes to the status information of all AGVs, and realizes global monitoring and path prediction. Based on the current position, speed and direction of the AGV, its future driving trajectory is predicted. Secondly, the conflict detection algorithm is used to calculate the possible collision points and times between AGVs. When it is predicted that the paths may intersect and the time is close, the conflict warning mechanism is triggered. Then, according to the preset priority strategy, the operating parameters of the relevant AGVs are dynamically adjusted, including the direction and docking position of the current AGV, and control instructions are sent to the AGV nodes, such as docking or re-planning the path, to avoid traffic congestion and collision events. Finally, based on factors such as the urgency of the task, the type of empty load of the AGV, and its own power status, the traffic priority of each AGV is determined to optimize the utilization of system resources.

[0042] When the scheduling module is called, the module is used to manage a dynamically updated task queue, which contains all tasks to be executed and their related information, such as task type, starting point, end point, expected completion time and priority; the task queue is sorted in real time according to priority, for example, according to factors such as the urgency, importance, expected benefits or resource requirements of the task, to ensure that high-priority tasks are given priority.

[0043] For each task to be executed, the scheduling module selects AGV vehicle nodes with corresponding capabilities and resources from the available AGV list for matching according to the specific requirements of the task, such as vehicle model requirements, load restrictions, power requirements, etc.; an intelligent matching algorithm is used to comprehensively consider the current position, status, remaining power, and task history factors of the AGV vehicle node to achieve the optimal match between the task and the AGV vehicle node.

[0044] For AGV vehicle nodes that have been assigned tasks, the scheduling module uses the A* algorithm to calculate the optimal path from the current position to the task endpoint; during the path calculation process, factors such as environmental constraints, obstacle distribution, path length, and number of turns are considered to ensure the feasibility and efficiency of the path; at the same time, the path planning results are combined with the dynamic characteristics of the AGV to smooth the path and generate an executable driving path.

[0045] During the AGV's task execution, the scheduling module continuously monitors the AGV's status changes and environmental dynamics, such as AGV failure, low battery, path congestion, etc. When an abnormal situation or environmental change is detected, the dynamic scheduling mechanism is immediately triggered to re-evaluate the task queue, AGV capabilities and path planning results, and make necessary adjustments and optimizations.

[0046] In this embodiment, the following is also performed: Establish a task execution feedback mechanism to collect real-time feedback information from AGV vehicle nodes during task execution, including but not limited to task progress, actual path execution status, power consumption, fault status, environmental perception data, and any abnormal events; conduct in-depth analysis and processing of the collected feedback information, and use data analysis algorithms and machine learning models to identify bottlenecks, inefficient links, and potential risk points in the task execution process; based on the analysis results of the feedback data, the scheduling module dynamically adjusts the task allocation strategy, path planning algorithm, and AGV scheduling rules for continuous optimization and adjustment; at the same time, each strategy adjustment based on feedback information is recorded, including performance comparison before and after the adjustment, specific measures for adjustment, implementation time, and adjustment effect evaluation, to provide reference and basis for subsequent task scheduling and path planning.

[0047] Specifically, the scheduling module of this embodiment performs task management and queue maintenance in advance, and receives task requests from the central control console or production system, which contain information such as the starting point, end point, empty load type and priority. The system maintains the queue of tasks to be executed and sorts them according to priority. Secondly, AGV capability evaluation and matching are performed to obtain the real-time status of all AGVs, including parameters such as position, power, and load capacity. According to the task requirements, AGVs with corresponding carrying capacity are matched to ensure the reliability of task execution. Then, path planning and optimization are performed, and the A* algorithm is used to calculate the optimal driving path of the AGV to avoid known obstacles and high traffic density areas. Considering dynamic environmental factors, the path is optimized in real time. Finally, a dynamic scheduling and feedback mechanism is established to update task allocation and path planning in real time when the AGV status or environment changes. By monitoring the execution of tasks, collecting feedback information from AGVs, and adjusting the scheduling strategy in a timely manner, the adaptability of the system is improved.

[0048] In this embodiment, when establishing a communication network, a functional analysis is performed on each ROS2 node in the system and classified according to the tasks and roles it undertakes; based on the functional classification, the nodes are modularly laid out, and nodes that perform similar or related tasks are placed in similar locations in the network topology to facilitate efficient exchange and management of data.

[0049] Assess the communication requirements of each ROS2 node, including data transmission volume, real-time requirements, and reliability requirements; allocate network bandwidth resources based on the communication demand assessment results to ensure that key nodes and communication links have sufficient bandwidth to meet data transmission requirements.

[0050] For node pairs that require high-frequency, large-volume communications, a point-to-point connection is used to reduce communication delays and packet loss rates.

[0051] By utilizing the multicast communication function of ROS2, for scenarios where the same data needs to be broadcast to multiple nodes, a multicast connection method is adopted to implement a one-to-many data transmission mode, and multiple AGV vehicle nodes receive the same control instructions or data updates at the same time.

[0052] The QoS configuration function of ROS2 is used to optimize the communication behavior of ROS2 nodes, including the configuration of parameters such as reliability, latency, and bandwidth limitation.

[0053] Establish a network performance monitoring system to monitor network status in real time, including key indicators such as bandwidth utilization, latency, and packet loss rate. Based on the monitoring results, regularly optimize network performance, such as adjusting node layout, optimizing path selection, and upgrading network equipment.

[0054] Implement encrypted communication strategies to encrypt transmission of sensitive data to ensure data security; design redundant communication paths and backup nodes to improve network reliability and fault tolerance; deploy firewalls and intrusion detection systems to prevent external attacks and network threats.

[0055] Establish a troubleshooting mechanism so that when a network failure occurs, the problem can be quickly located and repaired.

[0056] Specifically, the communication network of this embodiment uses the DDS communication mechanism of ROS2 in advance to configure the quality of service (QoS) parameters, such as reliability, delay, and historical record depth, to ensure the real-time and reliability of data transmission. Secondly, establish specific communication topics and services, implement data publish / subscribe and request / response mechanisms, define standard data message formats, and ensure compatibility between different nodes. Then, plan the network topology of ROS2 nodes, use multicast communication to reduce network load and improve communication efficiency; at the same time, monitor the network status, detect and handle abnormal situations such as communication delays and packet loss, and realize network topology design and optimization. Finally, through encryption and authentication means, ensure the security of communication data, prevent unauthorized access, set access rights for different nodes and users, protect the integrity of the system, and realize security mechanisms and rights management.

[0057] In this embodiment, when using the sensor data processing module, data from various sensor units of each AGV vehicle node is collected in real time, including but not limited to position information, speed information, acceleration information, obstacle detection information, ambient light information, temperature information, etc., and the collected data is transmitted to the central processing unit or the designated data processing node.

[0058] The collected raw data are preprocessed, including data cleaning, denoising, calibration, formatting, and time stamping.

[0059] Fuse the data from the sensor units of different AGV vehicle nodes, and conduct in-depth analysis and processing on the fused data; apply data fusion algorithms, such as multi-sensor data fusion, time series data fusion or spatial data fusion, to improve the fusion accuracy and reliability of data.

[0060] Conduct in-depth analysis and processing of the fused data, including but not limited to obstacle detection and recognition, path planning and analysis, traffic flow prediction, abnormal behavior detection, etc.

[0061] Apply machine learning or deep learning algorithms, such as cluster analysis, regression analysis, classification algorithms, or neural networks, to mine and model the fused data to extract valuable information and patterns; Based on the analyzed and processed data, environmental perception information is generated, including obstacle locations, traffic conditions, environmental anomalies, etc.; based on the environmental perception information, decision support suggestions are generated, such as path adjustment suggestions, traffic control instructions, emergency obstacle avoidance strategies, etc., to assist the traffic control module and the dispatching module to make smarter and more efficient decisions.

[0062] Specifically, the sensor data processing module of this embodiment uses the lidar sensor data to build a two-dimensional environmental map in advance, identify dynamic and static obstacles, and realize environmental perception and modeling; finally, the processed environmental information is published to the traffic control module and AGV node to support decision-making processes such as path planning, obstacle avoidance and traffic control, and realize data sharing and support decision-making.

[0063] Specifically, the central control console / human-computer interaction module of this embodiment develops a friendly user interface in advance to achieve real-time monitoring and visualization of the AGV operating status, location information, task progress and environmental map, and supports multi-perspective switching and data statistical display; secondly, it provides an interface for operators to input instructions, including task issuance, path modification, emergency stop and other functions, supports remote operation, and improves management flexibility; finally, when the system detects an abnormality or failure, it provides alarm prompts in various forms such as sound and vision, and allows operators to view system logs to understand detailed event information, thereby realizing alarm prompts and log viewing functions.

[0064] Specifically, the database / log module of this embodiment establishes an efficient database system in advance to store AGV operation logs, task records, sensor data and system events. By designing a reasonable data structure, it supports the storage and fast query of large amounts of data. Secondly, it provides statistical and analysis tools to mine and analyze historical data, generate performance reports and fault diagnosis reports, and support system optimization and maintenance. Finally, it regularly backs up the database to prevent data loss; it performs permission control and encryption processing on data access to ensure data security and privacy protection.

[0065] In this embodiment, when establishing an AGV vehicle node, an independent ROS2 node is configured for each AGV and the basic parameters of the ROS2 node are configured. Each AGV communicates and exchanges data as an independent entity in the network; wherein the basic parameters of the ROS2 node include the node name, namespace, communication protocol, etc., to ensure effective communication and data synchronization between nodes.

[0066] Integrate sensor units on the AGV, such as lidar, camera, odometer, etc., to obtain environmental perception data. Configure sensor parameters, such as scanning frequency, resolution, field of view, etc., to meet the needs of the SLAM algorithm.

[0067] Select and integrate SLAM algorithms suitable for AGV, such as Cartographer, Hector SLAM, Gmapping, etc.; configure SLAM algorithm parameters, such as map resolution, number of optimization iterations, feature extraction method, etc., to optimize SLAM performance.

[0068] The time synchronization mechanism of ROS2 is used to ensure that the data from different sensors remain consistent in time, and the data fusion algorithm is used to fuse the data from different sensors to improve the robustness and accuracy of environmental perception.

[0069] Start the SLAM algorithm and use sensor data to perform autonomous positioning and map construction of AGV in real time.

[0070] The constructed map is stored in the global map server of ROS2 for access and use by other nodes.

[0071] Conduct autonomous navigation tests of AGV in a real environment to verify the effectiveness and accuracy of the SLAM algorithm.

[0072] Specifically, the AGV vehicle node of this embodiment is to achieve autonomous navigation and positioning, and specifically includes the following contents: First, each AGV is configured as an independent ROS2 node, and the SLAM (simultaneous localization and mapping) algorithm is integrated to achieve autonomous positioning and mapping. By collecting and processing the laser radar sensor data, the positioning accuracy is improved to meet the centimeter-level accuracy requirements. Secondly, in terms of sensor deployment and data acquisition, a laser radar is installed on the AGV to obtain environmental information. The data collected by the laser radar sensor is published through the topic (Topics) of ROS2 for subscription and use by other modules to achieve data sharing. Thirdly, in order to achieve real-time obstacle avoidance and autonomous decision-making, a real-time obstacle avoidance algorithm is developed, which uses sensor data to detect obstacles ahead and dynamically adjusts the driving path of the AGV. Path planning and obstacle avoidance decisions are implemented locally to reduce dependence on the central system and improve response speed. Finally, the AGV node regularly publishes its own position information, speed, power and other status information, and provides real-time monitoring for the traffic control module and the scheduling module through the ROS2 communication mechanism to ensure the efficient operation of the system.

[0073] In this embodiment, the following is also performed: Continuously monitor nodes and the operating status of all AGV vehicle nodes and server nodes in real time, including but not limited to the node's connection status, processing power, memory usage, and communication delay.

[0074] Use the heartbeat mechanism or regular status reports to ensure that the status information of each node is updated in real time and fed back to the system monitoring center.

[0075] When a fault is detected in an AGV vehicle node or server node, the fault detection mechanism is immediately triggered to quickly locate and diagnose the fault.

[0076] Depending on the type and severity of the fault, the failover strategy is automatically executed, including switching to a pre-configured standby node, reallocating tasks to other normally operating AGVs, or taking other necessary emergency measures to ensure the continuous operation of the system.

[0077] After a failover, tasks are automatically reallocated and scheduled based on the current task status and AGV availability to ensure that all tasks are processed and completed in a timely manner.

[0078] In this embodiment, when the A* algorithm is used to calculate the optimal path from the current position to the task end point, it specifically includes: Initialize path search: determine the current position of the AGV as the starting point of the path search, and the end point of the task as the target position; initialize an open list Open List to store the nodes to be evaluated, and a closed list Closed List to store the nodes that have been evaluated; add the starting point node to the Open List, and set its cost to 0 and the parent node to be empty.

[0079] Define node cost and heuristic function: For each node, define its cost Cost as the actual path length from the starting point to the node; define a heuristic function to estimate the minimum cost from the current node to the target location, usually choosing Euclidean distance or Manhattan distance as the heuristic value.

[0080] Node evaluation: Select the node with the smallest sum of cost and heuristic value from the open list as the current node for evaluation; remove the current node from the open list and add it to the closed list, indicating that the node has been evaluated.

[0081] Expand the current node: Based on the position and environment information of the current node, generate all possible adjacent nodes, that is, the next position that can be moved to; for each adjacent node, calculate its cost as the cost of the current node plus the cost of moving from the current node to the adjacent node; if the adjacent node is in the closed list, ignore the node; otherwise, add it to the open list and set its parent node to the current node.

[0082] Update the open list: Sort the nodes in the open list to ensure that the node with the smallest sum of cost and heuristic value is always at the front of the list; repeat the steps of node evaluation and selection and expansion of the current node until the target position is added to the open list or the open list is empty.

[0083] Build the optimal path: If the target position is added to the open list, start from the target position and trace back to the starting point through its parent node to build the optimal path from the current position to the end point of the task; if the open list is empty and the target position is not found, it means that the target position cannot be reached, and an empty path or error message should be returned.

[0084] The constructed optimal path is smoothed to remove unnecessary turning points or redundant nodes to simplify the path and improve driving efficiency. Smoothing algorithms, such as line segment fitting, polynomial interpolation, or Bezier curves, are applied to smooth the path to generate a more continuous and smooth driving trajectory. Ensure that the smoothed path still meets the driving capabilities and safety constraints of the AGV, such as the maximum turning radius, minimum driving distance, etc.

[0085] During the driving process of the AGV, the path is dynamically adjusted according to the real-time updated environmental information and the current state of the AGV, and the adjusted path is sent to the actuator of the AGV to control the AGV to drive according to the adjusted path.

[0086] Through the above method steps, the AGV vehicle node uses the A* algorithm to calculate the optimal path from the current position to the task end point. This path planning method takes into account the actual path length and heuristic estimation value, which can improve the search efficiency while ensuring the optimality of the path. It enables the AGV to navigate autonomously in a complex and dynamic environment and reach the specified location of the task quickly and accurately.

[0087] In addition, in order to improve the fault tolerance and scalability of the system, the operating status of AGV nodes and server nodes is continuously monitored. When a node failure is detected, it can automatically switch to the backup node or reallocate the task to ensure the continuity of the system; secondly, redundant storage is used for key data to prevent data loss caused by single point failures, and security mechanisms are used to protect data integrity and reliability; finally, the system design supports the seamless integration of new AGVs and functional modules, and uses the distributed architecture of ROS2 to easily expand the system scale to meet the needs of different scenarios.

[0088] This embodiment provides a distributed AGV system based on ROS2, which includes an AGV vehicle node, a traffic control module, a scheduling module, a communication network, a sensor data processing module, a central control console / human-computer interaction interface, and a database / log module.

[0089] In practical applications, this embodiment provides a traffic control and scheduling method for a distributed AGV system based on ROS2, including the design of AGV vehicle nodes, the construction of traffic control modules, the development of scheduling modules, the establishment of communication networks, sensor data processing modules, central consoles / human-computer interaction interfaces, the establishment of database / log modules, and the fault tolerance and expansion capabilities of the system.

[0090] First, the design and implementation of the AGV vehicle node is the core component of this embodiment, which mainly includes three aspects: hardware composition, software architecture and autonomous decision-making. First, in terms of hardware composition, each AGV is equipped with a high-performance embedded computer running the ROS2 system to process complex sensor data and execute real-time control algorithms. In terms of sensor configuration, the AGV is equipped with a laser radar for environmental perception and obstacle detection, providing high-precision distance and angle information to help the AGV accurately identify obstacles and paths in the surrounding environment. The actuator includes a high-precision drive motor and a steering mechanism to achieve smooth movement and precise direction control of the AGV. The communication module supports Wi-Fi or a dedicated industrial wireless network to ensure real-time, high-speed communication between the AGV and other nodes and the central system, meeting the system's requirements for high efficiency and reliability of data exchange. Secondly, in terms of software architecture, ROS2 runs on the computing unit of each AGV, configuring it as an independent ROS2 node, making full use of the distributed architecture and efficient communication mechanism of ROS2. The SLAM (Synchronous Localization and Mapping) algorithm is integrated to achieve autonomous positioning and mapping of AGV in unknown or dynamic environments. In terms of navigation and control, the Nav2 navigation stack is integrated to achieve global path planning, local path planning and motion control. AGV can autonomously plan the driving path according to the global map and real-time environmental information, and make dynamic adjustments when necessary. A sensor driver has been developed to collect data from each sensor and publish it through ROS2 topics for subscription by the traffic control module, scheduling module and other AGV nodes to achieve real-time data sharing and collaborative work. In order to achieve real-time obstacle avoidance, an efficient obstacle avoidance algorithm has been developed using lidar data. AGV can detect static and dynamic obstacles in front in real time, dynamically adjust the driving path or speed to avoid collision risks. In terms of status information publishing, AGV nodes regularly publish their own location information, speed, power and other status information, and provide real-time monitoring for the traffic control module and scheduling module through ROS2 topics, supporting the global coordination and optimization of the system. Finally, in terms of autonomous decision-making, AGV has a high degree of autonomy and intelligence. After receiving the target position or task instruction, AGV uses a local planner for path planning, fully considering the current environment, road conditions and obstacle information, and calculates the optimal driving path. During driving, AGV continuously monitors the surrounding environment and uses real-time sensor data to respond immediately when encountering obstacles or emergencies, dynamically adjusting the path, slowing down or stopping to ensure the safety of itself and the surrounding environment. In addition, when network communication is interrupted or an emergency stop command is received, AGV can autonomously handle safety. For example, if communication with the central system is interrupted, AGV will choose to safely stop at the current location or return to the nearest safe area according to the preset safety strategy and wait for communication to be restored.This autonomous decision-making capability reduces the dependence on the central system, improves the robustness and fault tolerance of the system, and ensures safe operation under various abnormal conditions. Through the above design and implementation, the AGV vehicle node has high-precision environmental perception capabilities, intelligent autonomous decision-making capabilities, and reliable communication and coordination capabilities, laying a solid foundation for the efficient operation of the distributed AGV system and the implementation of traffic control and scheduling methods.

[0091] Furthermore, the construction of the traffic control module is the key to ensure the safe and efficient operation of the multi-AGV system. First, in terms of global monitoring and path prediction, it is necessary to deploy traffic control nodes on the central server, run the ROS2 system, and subscribe to the status information of all AGV nodes. By collecting the location information, speed and direction of each AGV, a global AGV status database is established to provide comprehensive and real-time dynamic data support for the system. Based on these data, the AGV path prediction algorithm is used to predict the future driving path trajectory of the AGV and predict the possible traffic conditions in advance. Next, in the conflict detection and early warning link, the traffic control module uses time and space parameters to calculate the time and location where the AGV may conflict. Advanced conflict detection algorithms, such as spatiotemporal collision detection models, are used to quickly identify potential collision risks. When a potential conflict is detected, the early warning mechanism is immediately triggered, the relevant AGVs are marked, and the dispatcher and the AGV itself are notified to ensure timely measures. In terms of coordinated control and command issuance, firstly, the traffic priority of each AGV is determined according to the urgency of the task, the load type and the current status of the AGV, and a reasonable avoidance strategy is formulated. Generate specific control instructions for the AGV that needs to be avoided, including operations such as slowing down, docking, or re-planning the path. Through the ROS2 service (Services) or action (Actions) interface, the control instructions are sent to the relevant AGV in real time to ensure the rapid transmission and execution of the instructions. Finally, in the feedback and adjustment stage, after executing the instructions, the AGV promptly feedbacks the execution results, such as speed changes, position updates, or path adjustments. Based on these feedbacks, the traffic control module updates the global AGV status database and continuously optimizes the traffic control strategy. Based on the latest AGV status and environmental information, the control parameters are dynamically adjusted to ensure the overall coordination and operation efficiency of the system. Through the collaborative work of the above modules, the traffic control module effectively prevents conflicts and traffic congestion between AGVs, improves the safety and reliability of the multi-AGV system, and ensures the smooth progress of logistics transportation.

[0092] Furthermore, the development of the scheduling module is a key link in optimizing the efficiency and coordination of the AGV system. First, in terms of task management and queue maintenance, the scheduling module receives task requests from the central console or production system, which contain detailed information such as the starting point, end point, load type and priority. The module adds these tasks to the queue of tasks to be executed and sorts the queue according to the priority and deadline of the tasks to ensure that high-priority and urgent tasks can be processed in time. Secondly, the scheduling module evaluates and matches the AGV capabilities. The status information of all AGVs is obtained in real time, including location information, power level, load capacity and current operating status. According to the task requirements, eligible AGVs are screened and their availability and applicability are evaluated. For example, for tasks that require the transportation of heavy materials, the system will select AGVs with high load capacity and sufficient power to ensure the reliability and efficiency of task execution. In terms of path planning and optimization, the scheduling module uses a global path planning algorithm, such as the A* algorithm, to calculate the optimal driving path for the selected AGV. Real-time traffic control information is taken into account during the planning process to avoid known obstacles, congested areas and high-risk sections, thereby improving the safety and efficiency of operation. In addition, the module supports dynamic avoidance function, which can re-plan the path in real time according to the changes in environment and traffic conditions to ensure that the AGV can complete the task smoothly. Finally, in the task allocation and dynamic scheduling stage, the scheduling module sends the task instructions to the matching AGV node. The module has dynamic scheduling capabilities. When the AGV status (such as low power, failure) or environment changes, it can adjust the task allocation and path planning in real time, reselect the appropriate AGV or modify the driving route. At the same time, a feedback mechanism is established to monitor the execution of tasks and collect feedback information from AGVs, such as task progress, position deviation and operation abnormalities. Based on these feedbacks, the scheduling module timely optimizes the scheduling strategy, adjusts the parameter settings, and improves the adaptability and overall operation efficiency of the system. Through the integration of the above functions, the scheduling module realizes the full process optimization of task management, AGV matching, path planning and dynamic scheduling, ensuring that the AGV system operates efficiently and reliably in a dynamic and complex industrial environment. This not only improves the efficiency and accuracy of material handling, but also effectively reduces the operating cost and maintenance difficulty of the system.

[0093] Furthermore, the sensor data processing module plays a key role in the system, mainly including two parts: environmental perception and modeling, and data sharing and decision support. First, in terms of environmental perception and modeling, the module uses the SLAM (Simultaneous Localization and Mapping) algorithm to generate a two-dimensional map of the environment and achieve accurate modeling of the work area. This provides a reliable foundation for the autonomous navigation and positioning of the AGV. At the same time, the system uses clustering and classification algorithms to conduct in-depth analysis of the collected sensor data to identify and distinguish dynamic and static obstacles. Such obstacle detection capabilities enable the AGV to perceive environmental changes in real time, make avoidance or adjustments in time, and improve the safety and efficiency of operation. Secondly, in terms of data sharing and decision support, the processed environmental information is published through ROS2 topics (Topics) for subscription by the traffic control module and other AGV nodes. This efficient data sharing mechanism ensures information synchronization and collaborative work among various parts of the system. Accurate environmental data provides strong support for decision-making processes such as path planning, real-time obstacle avoidance and traffic control, helping the system maintain optimal performance in a dynamic and complex environment. Through continuous data updating and sharing, the system can respond to environmental changes in a timely manner, optimize the operation strategy of the AGV, and improve the overall level of intelligence. In summary, the sensor data processing module enhances the system's environmental perception capability and decision support function through algorithms and efficient data transmission, providing important guarantees for the efficient operation and safety of the distributed AGV system.

[0094] Furthermore, the database / log module plays a vital role in the distributed AGV system of this embodiment, responsible for efficient data storage, management and analysis, and ensuring the stability and optimization capability of the system operation. First, in terms of data storage and structural design, the system uses the relational database MySQL to store various types of data, and the specific choice depends on the degree of data structuring and access requirements. In order to improve the efficiency of data storage and query performance, a reasonable data table structure is designed, and necessary indexes are established. In addition, considering the large-scale data that may be generated by the system, partitioning and sharding strategies are implemented to support the storage and rapid access of massive data, ensuring that the database can still operate efficiently in a high-concurrency environment. Secondly, in terms of data analysis and report generation, the module integrates the advanced data analysis tool Spark to conduct in-depth mining and analysis of historical data. These tools can process and analyze large amounts of data, extract valuable information, and help identify trends and potential problems in system operation. Through the automated data analysis process, the system can generate detailed performance reports and fault diagnosis reports, which support multiple export formats (such as PDF, Excel) for easy review by operators and managers. This not only improves the maintenance efficiency of the system, but also provides a scientific basis for further system optimization. In terms of data backup and security management, the module has designed a complete data backup strategy to ensure the security and integrity of the data. The system has set up an automatic backup mechanism to support full backup and incremental backup to prevent data loss or damage due to unexpected circumstances. In order to ensure the security of data, strict permission control is implemented, and fine-grained permission management is performed on database access to ensure that only authorized personnel can perform data operations. In addition, sensitive data is stored and transmitted through encryption technology to prevent unauthorized access or tampering during storage and transmission. Through these measures, the database / log module not only achieves efficient data management and fast access, but also ensures data security and privacy protection, and improves the reliability and trust of the entire system.

[0095] Furthermore, the database / log module fully supports the efficient operation and continuous optimization of the distributed AGV system through scientific data storage structure design, advanced data analysis tool integration, and strict data backup and security management, ensuring the stability and efficiency of the system in complex industrial environments.

[0096] In summary, the present invention has a high degree of fault tolerance and expansion capability to ensure continuous and stable operation in a complex industrial environment. First, through the AGV node monitoring and fault switching mechanism, the heartbeat detection technology is used to monitor the operating status of each AGV node and server node in real time. Once a node is detected to have a fault, the fault switching program can be started immediately, automatically switching to the standby node or reallocating the task to ensure the continuity and reliability of the system. In addition, in terms of data redundancy and security, multiple copies of key data are stored, and the use of distributed databases and file systems significantly improves the availability and fault tolerance of data, preventing data loss or system shutdown caused by single point failures. At the same time, a modular architecture is designed to facilitate adding functional modules or upgrading existing modules without affecting the operation of the existing system, so as to achieve flexible system expansion. By supporting dynamic expansion, new AGVs or service nodes can be dynamically added without interrupting operation to meet application scenarios of different scales and requirements. These designs and implementation measures jointly construct a distributed AGV system with high fault tolerance and good scalability, which can effectively deal with various faults and changes, and ensure that the system always maintains an efficient and stable operating state in a changing industrial environment.

[0097] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, 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 specification.

[0098] The above-mentioned embodiments are only preferred embodiments of the present invention and cannot be used to limit the scope of protection of the present invention. Any non-substantial changes and substitutions made by technicians in this field on the basis of the present invention shall fall within the scope of protection required by the present invention.

Claims

1. A traffic control and dispatching method for a distributed AGV system based on ROS2, characterized in that: The following steps are involved: Establish AGV vehicle nodes: configure each AGV as an independent ROS2 node, and integrate the synchronous positioning and mapping algorithm to achieve autonomous positioning and mapping; install a sensor unit on each AGV to obtain surrounding environment information; each AGV regularly publishes its own status information through the ROS2 communication mechanism; Constructing traffic control module: Establishing a traffic control node on the central server as a global traffic monitoring center; this node subscribes to the status information of all AGV vehicle nodes and implements conflict detection algorithms to predict potential traffic conflicts; when potential conflicts are detected, control instructions are sent to relevant AGV vehicle nodes; at the same time, priority strategies are formulated based on priority factors to determine the traffic priority of each AGV vehicle node; Develop a scheduling module: receive and manage task requests from different sources; obtain the real-time status of all AGV vehicle nodes and evaluate their availability and applicability; use the path planning algorithm to calculate the shortest driving path / optimal driving path for the AGV vehicle nodes; assign AGVs to perform tasks based on the status of the AGV vehicle nodes, task requirements, and path planning results, and send corresponding task instructions; Establish a communication network: Use the DDS communication mechanism of ROS2 to realize data exchange and communication between AGV vehicle nodes, traffic control modules and scheduling modules; Integrated sensor data processing module: collects and processes data from the sensor units of AGV vehicle nodes, integrates, analyzes and processes the collected data, and provides environmental perception and decision support for the traffic control module and the dispatching module; Provide central control console / human-computer interaction interface: Design and implement a central control console or human-computer interaction interface to display the AGV's operating status, task progress, and system alarm information; Establish database / log module: Establish a database system to store historical task data, AGV operation logs and system events, and provide data query, analysis and export functions.

2. The method according to claim 1, characterized in that: When implementing the conflict detection algorithm, by continuously subscribing to the ROS2 topics published by all AGV vehicle nodes, the location information, operating status, current task endpoint and travel direction of each AGV vehicle node are received and updated in real time; Based on the current position, speed, direction and task destination of each AGV vehicle node, a path planning algorithm is used to predict its path trajectory in the future. At the same time, a conflict detection algorithm is implemented to compare the spatial position relationship of each AGV predicted path to identify possible path intersections or approaches, that is, potential collision risks. When the conflict detection algorithm predicts that the paths of any two or more AGV vehicle nodes will intersect or approach to within a preset safety distance at a certain time point or spatial area in the future, the conflict warning mechanism is immediately triggered; the warning information includes the type of conflict, the expected time and location, the AGV numbers involved, and the recommended avoidance measures; According to the preset priority strategy and rules, the AGV vehicle nodes that need to adjust the path are determined; the dynamic path replanning algorithm is used to calculate new feasible paths for the affected AGV vehicle nodes; The calculated new path is sent to the relevant AGV vehicle node in the form of ROS2 service or action message, instructing it to drive along the new path; at the same time, the execution of the adjustment instruction by the AGV vehicle node is continuously monitored, including the path tracking accuracy, speed control and the implementation effect of the avoidance measures.

3. The method according to claim 1, characterized in that: When the scheduling module is called, the module is used to manage a dynamically updated task queue, which contains all tasks to be executed and their related information; the task queue is sorted in real time according to priority; For each task to be executed, the scheduling module selects an AGV vehicle node with corresponding capabilities and resources from the available AGV list for matching according to the specific requirements of the task; an intelligent matching algorithm is used to comprehensively consider the current position, status, remaining power, and task history factors of the AGV vehicle node to achieve the optimal matching between the task and the AGV vehicle node; For AGV vehicle nodes that have been assigned tasks, the scheduling module uses the A* algorithm to calculate the optimal path from the current position to the task end point; at the same time, the path planning results are combined with the dynamic characteristics of the AGV to smooth the path and generate an executable driving path; During the AGV's task execution, the scheduling module continuously monitors the AGV's status changes and environmental dynamics; when an abnormal situation or environmental change is detected, the dynamic scheduling mechanism is immediately triggered to re-evaluate the task queue, AGV capabilities, and path planning results, and make necessary adjustments and optimizations.

4. The method according to claim 3, characterized in that Also execute: Establish a task execution feedback mechanism to collect feedback information from AGV vehicle nodes in real time during task execution; conduct in-depth analysis and processing of the collected feedback information, and use data analysis algorithms and machine learning models to identify bottlenecks, inefficient links, and potential risk points in the task execution process; Based on the analysis results of the feedback data, the scheduling module dynamically adjusts the task allocation strategy, path planning algorithm and AGV scheduling rules for continuous optimization and adjustment; at the same time, each strategy adjustment based on feedback information is recorded, including the performance comparison before and after the adjustment, the specific measures of the adjustment, the implementation time and the evaluation of the adjustment effect, to provide reference and basis for subsequent task scheduling and path planning.

5. The method according to claim 1, characterized in that: When establishing a communication network, perform a functional analysis on each ROS2 node in the system and classify them according to the tasks and roles they undertake. According to the functional classification, the nodes are modularly laid out, and nodes that perform similar or related tasks are placed in close locations in the network topology. Evaluate the communication requirements of each ROS2 node, including data transmission volume, real-time requirements, and reliability requirements; allocate network bandwidth resources based on the communication demand assessment results to ensure that key nodes and communication links have sufficient bandwidth; For node pairs that require high-frequency, large-volume communications, a point-to-point connection is used to reduce communication delays and packet loss rates; Utilizing the multicast communication function of ROS2, for scenarios where the same data needs to be broadcast to multiple nodes, a multicast connection method is used to implement a one-to-many data transmission mode, where multiple AGV vehicle nodes receive the same control instructions or data updates at the same time. The QoS configuration function of ROS2 is used to optimize the communication behavior of ROS2 nodes.

6. The method according to claim 1, characterized in that: When using the sensor data processing module, data from various sensor units of each AGV vehicle node is collected in real time, and the collected data is transmitted to the central processing unit; Preprocess the collected raw data, including data cleaning, denoising, calibration, formatting, and time stamping; Fuse the data from the sensor units of different AGV vehicle nodes, and conduct in-depth analysis and processing on the fused data; Apply machine learning or deep learning algorithms to mine and model fused data to extract valuable information and patterns; Based on the analyzed and processed data, environmental perception information is generated, decision support suggestions are generated, and real-time and accurate environmental descriptions are provided for traffic control modules and scheduling modules.

7. The method according to claim 1, characterized in that: When establishing an AGV vehicle node, configure an independent ROS2 node for each AGV and configure the basic parameters of the ROS2 node. Each AGV communicates and exchanges data as an independent entity in the network. Integrate the sensor unit on the AGV to obtain environmental perception data; Select and integrate the SLAM algorithm suitable for AGV and configure the SLAM algorithm parameters; The time synchronization mechanism of ROS2 is used to ensure that the data from different sensors are consistent in time, and the data from different sensors are fused through the data fusion algorithm; Start the SLAM algorithm and use sensor data to perform autonomous positioning and map construction of the AGV in real time; The constructed map is stored in the global map server of ROS2 for access and use by other nodes.

8. The method according to claim 1, characterized in that Also execute: Continuously monitor nodes and monitor the operating status of all AGV vehicle nodes and server nodes in real time; Use the heartbeat mechanism or regular status reports to ensure that the status information of each node is updated in real time and fed back to the system monitoring center; When a fault is detected in an AGV vehicle node or server node, the fault detection mechanism is immediately triggered to quickly locate and diagnose the fault; Automatically execute failover strategies based on the fault type and severity, including switching to pre-configured backup nodes and reallocating tasks to other normally operating AGVs; After a failover, tasks are automatically reallocated and scheduled based on the current task status and AGV availability to ensure that all tasks are processed and completed in a timely manner.

9. The method according to claim 3, characterized in that: When using the A* algorithm to calculate the optimal path from the current location to the task destination, it specifically includes: Initialize path search: determine the current position of the AGV as the starting point of the path search, and the task end point as the target position; initialize an open list Open List to store the nodes to be evaluated, and a closed list ClosedList to store the nodes that have been evaluated; add the starting point node to the Open List, and set its cost to 0 and the parent node to be empty; Define node cost and heuristic function: For each node, define its cost as the actual path length from the starting point to the node; define a heuristic function to estimate the minimum cost from the current node to the target position; Node evaluation: Select the node with the smallest sum of cost and heuristic value from the open list as the current node for evaluation; remove the current node from the open list and add it to the closed list, indicating that the node has been evaluated; Expand the current node: Based on the position and environment information of the current node, generate all possible adjacent nodes, that is, the next position that can be moved to; for each adjacent node, calculate its cost as the cost of the current node plus the cost of moving from the current node to the adjacent node; if the adjacent node is in the closed list, ignore the node; otherwise, add it to the open list and set its parent node to the current node.

10. The method according to claim 9, characterized in that: Update the open list: sort the nodes in the open list to ensure that the node with the smallest sum of cost and heuristic value is always at the front of the list; repeat the steps of node evaluation and selection and expansion of the current node until the target position is added to the open list or the open list is empty; Construct the optimal path: If the target location is added to the open list, start from the target location and trace back to the starting point through its parent node step by step to construct the optimal path from the current location to the end point of the task; if the open list is empty and the target location is not found, it means that the target location cannot be reached, and an empty path or error message should be returned at this time; Smoothing the constructed optimal path and removing unnecessary turning points or redundant nodes to simplify the path and improve driving efficiency; During the driving process of the AGV, the path is dynamically adjusted according to the real-time updated environmental information and the current state of the AGV, and the adjusted path is sent to the actuator of the AGV to control the AGV to drive according to the adjusted path.

Images