Path planning method and system of AGV (Automatic Guided Vehicle) for silver electrolysis process workshop

Abstract

The invention relates to the technical field of path planning, and discloses a path planning method and system of an AGV (Automatic Guided Vehicle) for a silver electrolysis process workshop. The method comprises the following steps: carrying out environment characteristic acquisition and electronic map construction on a silver electrolysis process workshop to obtain a silver electrolysis process workshop electronic map; calculating an initial optimal path of the AGV from the current position to the target workstation; detecting the actual driving resistance in the execution process of the initial optimal path to obtain a path correction parameter; performing local adjustment to obtain an energy-saving safe path; and in the process that the AGV executes the energy-saving safe path, the front moving obstacle and the electrolyte leakage area are monitored in real time, and a temporary detour path is generated. According to the application, the path planning is more suitable for the actual environment characteristics of the silver electrolysis workshop, the emergency situation of the silver electrolysis workshop can be timely handled, the equipment safety and the production continuity are guaranteed, the no-load rate of the AGV is reduced, and the overall logistics efficiency of the silver electrolysis workshop is improved.

Description

Technical Field

[0001] The present application relates to the field of path planning technology, and in particular to a path planning method and system for an AGV transport vehicle used in a silver electrolysis process workshop. Background Art

[0002] Silver electrolysis, a key process in nonferrous metal smelting, operates in a highly corrosive environment characterized by densely packed equipment and complex process flows. Traditional manual handling methods are not only inefficient but also pose significant safety risks. Conventional automated transport vehicles (AMTs) struggle to adapt to the unique environmental requirements of silver electrolysis workshops. Electrolytic cells are densely packed, and materials such as cathode and anode plates require regular replacement and handling. Furthermore, the complex and changing floor conditions and uneven distribution of corrosive gases make conventional path planning methods difficult to meet practical application requirements.

[0003] Existing AGV (Automated Guided Vehicle) (AGV) path planning methods are mostly designed for standard industrial environments and fail to consider factors unique to silver electrolysis workshops, such as the corrosive environment, irregular equipment layout, and dynamically changing ground conditions. When applied in silver electrolysis workshops, these methods often suffer from problems such as irrational path selection, frequent detours, and excessive energy consumption. Path planning accuracy and stability are particularly degraded when floor conditions and workloads vary, leading to low handling efficiency. Furthermore, conventional path planning algorithms are unable to effectively address emergencies such as electrolyte leaks, increasing the risk of equipment damage and production interruptions. When multiple AGVs work together in a silver electrolysis workshop, existing technologies struggle to achieve efficient task allocation and path coordination. This is especially true when handling cathode and anode plates simultaneously, where path conflicts and resource contention are common. Traditional centralized scheduling systems are slow to respond and struggle to cope with dynamic obstacles and unexpected situations within the workshop, while distributed systems lack overall coordination capabilities and struggle to optimize global resource allocation. Furthermore, existing technologies fail to fully consider the unique process flow and task priorities of silver electrolysis, reducing overall production efficiency and impacting process stability. Summary of the Invention

[0004] The present application provides a path planning method and system for AGV transport vehicles in a silver electrolysis process workshop. The present application makes path planning more in line with the actual environmental characteristics of the silver electrolysis workshop, can respond to emergencies in the silver electrolysis workshop in a timely manner, ensure equipment safety and production continuity, reduce the empty load rate of AGV transport vehicles, and improve the overall logistics efficiency of the silver electrolysis workshop.

[0005] In a first aspect, the present application provides a path planning method for an AGV transport vehicle in a silver electrolysis process workshop, the path planning method for an AGV transport vehicle in a silver electrolysis process workshop comprising:

[0006] Collect environmental characteristics and construct an electronic map of the silver electrolysis process workshop to obtain an electronic map of the silver electrolysis process workshop;

[0007] Calculate the initial optimal path of the AGV transport vehicle from the current position to the target workstation according to the electronic map of the silver electrolysis process workshop;

[0008] detecting the actual driving resistance during the execution of the initial optimal path, and when the abnormal resistance value of a path segment exceeds a momentum residual threshold, recording the actual friction coefficient and obstacle conditions of the path segment to obtain a path correction parameter;

[0009] According to the path correction parameters, the initial optimal path is locally adjusted by bypassing high corrosion areas and high friction coefficient areas to obtain an energy-saving and safe path;

[0010] When the AGV transport vehicle executes the energy-saving and safe path, it monitors the moving obstacles and electrolyte leakage area in front in real time, calculates the temporary obstacle avoidance direction and speed, and generates a temporary detour path.

[0011] In a second aspect, the present application provides a path planning system for an AGV transport vehicle in a silver electrolysis process workshop, the path planning system for an AGV transport vehicle in a silver electrolysis process workshop comprising:

[0012] A construction module is used to collect environmental characteristics of the silver electrolysis process workshop and construct an electronic map to obtain an electronic map of the silver electrolysis process workshop;

[0013] A calculation module, configured to calculate an initial optimal path for the AGV transport vehicle from a current position to a target workstation based on an electronic map of the silver electrolysis process workshop;

[0014] a detection module, configured to detect actual driving resistance during execution of the initial optimal path, and when an abnormal resistance value of a path segment exceeds a momentum residual threshold, record the actual friction coefficient and obstacle conditions of the path segment, and obtain a path correction parameter;

[0015] A local adjustment module is used to locally adjust the initial optimal path by bypassing high corrosion areas and high friction coefficient areas according to the path correction parameters to obtain an energy-saving and safe path;

[0016] The real-time monitoring module is used to monitor the moving obstacles and electrolyte leakage areas ahead in real time while the AGV transport vehicle is executing the energy-saving and safe path, and calculate the temporary obstacle avoidance direction and speed to generate a temporary detour path.

[0017] The technical solution provided in this application accurately distinguishes electrolytic cell areas, corrosive gas areas, and equipment-intensive areas through the collection of environmental characteristics of the silver electrolysis workshop and the construction of a refined electronic map. This provides detailed environmental information to the AGV, enabling path planning to better align with the actual environmental characteristics of the silver electrolysis workshop. A comprehensive scoring mechanism is used to calculate the initial optimal path, taking into account path length, exposure to the corrosive environment, and equipment collision risk. This reduces the AGV's dwell time in high-corrosion areas, extending equipment life and reducing maintenance costs. Dynamic detection of the actual driving resistance along the path and recording of correction parameters enable real-time adaptation to changes in the floor conditions of the silver electrolysis workshop, making the path planning solution more suitable for the complex and changing actual environment of the silver electrolysis workshop. Local adjustments are made based on the path correction parameters to bypass high-corrosion and high-friction areas, reducing energy consumption and improving the operating efficiency and safety of the AGV in the silver electrolysis workshop. Real-time monitoring of moving obstacles and electrolyte leak areas ahead allows calculation of temporary obstacle avoidance directions and speeds, enabling the AGV to respond promptly to emergencies in the silver electrolysis workshop, ensuring equipment safety and production continuity. Through the binary task structure and leader-follower architecture, multiple AGV transporters can work together, optimize the transportation process of cathode and anode plates, reduce the empty load rate of AGV transporters, and improve the overall logistics efficiency of the silver electrolysis workshop. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0019] Figure 1 This is a schematic diagram of an embodiment of a path planning method for an AGV transport vehicle in a silver electrolysis process workshop according to an embodiment of the present application;

[0020] Figure 2 This is a schematic diagram of an embodiment of a path planning system for an AGV transport vehicle used in a silver electrolysis process workshop in an embodiment of the present application. DETAILED DESCRIPTION

[0021] The embodiments of the present application provide a path planning method and system for an AGV transport vehicle for a silver electrolysis process workshop. The terms "first", "second", "third", "fourth", etc. (if any) in the specification and claims of this application and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms "including" or "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or that are inherent to these processes, methods, products or devices.

[0022] For ease of understanding, the specific process of the embodiment of the present application is described below. Figure 1 In one embodiment of the present application, a path planning method for an AGV transport vehicle in a silver electrolysis process workshop includes:

[0023] Step S101: collecting environmental characteristics of a silver electrolysis process workshop and constructing an electronic map to obtain an electronic map of the silver electrolysis process workshop;

[0024] It is understandable that the execution subject of this application can be a path planning system for an AGV transport vehicle used in a silver electrolysis process workshop, or a terminal or a server, and the specific implementation is not limited here. The embodiment of this application is described by taking the server as the execution subject as an example.

[0025] Specifically, data is collected based on the physical layout of the silver electrolysis workshop, including information on the location of static obstacles, equipment layout, process pipeline locations, and the distribution of work areas. By deploying multiple high-precision positioning camera systems within the workshop, combined with LiDAR scanning and visual recognition systems, the spatial structure of the workshop is captured. This includes key information such as equipment installation areas, aisle layout, electrolytic cell locations, and storage areas. This also includes the layout of the pipeline network, which poses a potential obstacle to the AGV's travel path. The raw workshop environmental data is input into the system for rasterization, dividing the entire space of the silver electrolysis workshop into equal-sized basic cells, each representing a physical area. The cell size is set to 10 cm x 10 cm to ensure sufficient spatial resolution for path planning. Each cell is assigned different attribute tags, which distinguish different functional areas of the workshop: free areas, static obstacles, electrolytic cell areas, cathode plate storage areas, anode plate storage areas, short-circuit plate storage areas, aisle areas, and areas with high corrosive gas concentrations. Free zones indicate free passage for AGVs, while static obstacle zones indicate areas that are inaccessible or contain immovable obstacles. The electrolytic cell area, a critical process area in the workshop, is marked as a special area to prevent accidental entry and potential hazards. Cathode and anode plate storage areas are key locations for AGVs to perform their tasks, and path planning for these areas requires priority optimization. Short-circuit plate storage areas and other storage areas also require clear identification. Areas with high concentrations of corrosive gases are specially marked so that path planning can proactively avoid these areas, extending the lifespan of AGVs and reducing the risk of equipment damage. The marked raster data is converted into a workshop topology map. A topology map is the core form of an electronic map. Each grid cell is mapped to a node in the topology map. These nodes represent accessible locations and are connected by edges. Edges are established based on the physical location relationships between grid cells. Connecting edges are established between adjacent cells, and these edges are assigned weights based on the characteristics of different areas. The edge weights are determined by multiple factors, including the actual path distance, the corrosiveness of the area, and the friction coefficient of the ground. The resulting weighted edge set provides path information for the subsequent path planning algorithm, enabling the system to select the optimal path based on the characteristics of different areas. Based on this weighted edge set, all workstations are marked to construct an electronic map of the silver electrolysis process workshop. Workstations marked include the electrolytic cell plate removal station, cathode plate storage station, and anode plate storage station.

[0026] Step S102: Calculate the initial optimal path of the AGV from the current position to the target workstation based on the electronic map of the silver electrolysis process workshop;

[0027] Specifically, the AGV's starting position and destination workstation are determined based on its real-time location and task requirements within the silver electrolysis process workshop. This process involves analyzing the current task status, including determining whether the task is to transport cathode or anode plates. The system then obtains the corresponding destination workstation coordinates from the task management system, accurately locating the starting and ending coordinates. This coordinate information forms the basis for path planning. Based on the starting and ending coordinates, the starting and ending points of the path search are marked on an electronic map, providing accurate input parameters for the subsequent path search algorithm. Based on the starting and ending coordinates, a priority queue is constructed on the electronic map of the silver electrolysis process workshop. The starting node is added to the priority queue, and the path cost is initialized to obtain the initial state of the path search. During the path search, the priority queue is used to store pending nodes. The system always prioritizes expanding nodes with the lowest cost to ensure that the best paths are considered during the path search. As the search progresses, the system continuously selects nodes with the lowest cost from the priority queue for expansion and calculates the path costs of their adjacent nodes to obtain a set of candidate path nodes. Nodes in the candidate path set are assigned different travel cost coefficients based on the type of area they are located in to better reflect the travel difficulty in different areas. Due to the complex environment of the silver electrolysis plant, travel costs vary significantly across different areas. Different travel costs are therefore set for different areas. For example, the corridor area has the lowest travel cost coefficient, while the electrolysis tank area, due to its narrow space and higher risk, has a higher travel cost coefficient. A turning penalty factor is introduced into the path cost calculation to adjust the path cost to account for the additional cost of turning on the AGV. The turning penalty factor is calculated based on the angle between adjacent path segments; larger angles increase the penalty factor. This minimizes unnecessary turns during the path search process and optimizes overall path smoothness and safety. Based on the adjusted node cost, each candidate path is safety-verified to ensure that all paths meet the safety requirements of the silver electrolysis plant. This safety verification checks the minimum safe distance between the path and static obstacles, and also considers whether the path passes through high-corrosion areas and high-friction areas that may cause equipment damage. Paths that meet safety constraints are retained to form a path set. A comprehensive score is calculated based on a set of paths that meet safety constraints. Scoring criteria include path length, driving cost, energy consumption, number of turns, and safety. Weights are weighted for each dimension, and a comprehensive score is calculated for each path. The path with the highest score from the set of paths that meet safety constraints is selected as the initial optimal path and provided to the AGV for execution.

[0028] Step S103: Detecting the actual driving resistance during the execution of the initial optimal path. When the abnormal resistance value of a path segment exceeds the momentum residual threshold, record the actual friction coefficient and obstacle conditions of the path segment to obtain a path correction parameter.

[0029] Specifically, a multi-sensor system installed on the AGV collects real-time driving data and obtains raw data on the vehicle's status during travel. These sensors include distance sensors, a visual recognition system, and torque sensors. The distance sensors detect surrounding obstacles and path conditions, the visual recognition system identifies path landmarks, equipment status, and regional features, and the torque sensors detect changes in the vehicle's applied force during travel. Through the collaborative operation of these sensors, real-time driving data for each path segment is acquired as the AGV executes its path, providing environmental parameters and status information for subsequent path corrections. The raw driving data of the AGV's status is segmented by path segment, dividing the driving process into multiple path segments. Each path segment corresponds to a travel interval on the initial optimal path. The segmented driving parameters include travel time, average speed, applied force, and other key data for each path segment. Based on the segmented driving data, the momentum residual value is calculated for each path segment. The momentum residual determines whether the path segment resistance is abnormal by comparing the theoretical momentum with the actual momentum change. The calculation of momentum residual is based on the mass, average speed and force-time product of the AGV transport vehicle. The calculated momentum residual data reflects the change in actual resistance during the transport vehicle's driving process. If the momentum residual value of a path segment is high, it means that the resistance of the path segment is abnormal. In order to judge the abnormality of the path segment, the momentum residual data of the path segment is compared with the preset momentum residual threshold. When the momentum residual value exceeds the momentum residual threshold, the path segment is marked as an abnormal path segment, and a set of abnormal path segments is obtained. The adaptive learning rate is calculated based on the momentum residual value of the abnormal path segment to obtain the parameter update coefficient. The adaptive learning rate is a dynamically adjusted parameter that automatically adjusts the update step size according to the size of the momentum residual value of the path segment, thereby effectively correcting the friction coefficient of the abnormal path segment. Based on the calculated parameter update coefficient, the ground friction coefficient of each path segment in the abnormal path segment set is corrected and calculated, and the corrected friction coefficient is used as the path correction parameter.

[0030] Step S104: Based on the path correction parameters, the initial optimal path is locally adjusted by bypassing high corrosion areas and high friction coefficient areas to obtain an energy-saving and safe path;

[0031] Specifically, path correction parameters are fed back into the electronic map of the silver electrolysis process workshop to modify the edge weights of relevant areas. These path correction parameters are obtained after detecting abnormal resistance in a path segment and calculating the actual friction coefficient. These parameters are used to dynamically adjust the weights of each path segment in the electronic map, ensuring that the weight data reflects the current actual environmental conditions in the workshop. Based on the correction parameters, the weighted edge set of the silver electrolysis process workshop is updated, appropriately increasing the edge weights of areas with high corrosion and high friction coefficients. This allows path planning to prioritize avoiding these areas during subsequent adjustments, resulting in updated electronic map weight data. Based on this updated electronic map weight data and the initial optimal path, a set of path segments to be adjusted is identified. These path segments include those with abnormal momentum residuals, those in high corrosion areas, and those with high friction coefficients. These path segments affect the energy consumption and driving safety of the AGV and may cause the vehicle to stall or equipment damage. By analyzing the cost value of the path segment set, path segments with higher weights are prioritized and targeted for adjustment. Each path segment in the set of path segments to be adjusted is smoothed to bypass high-corrosion and high-friction areas, resulting in a locally optimized path segment. Path smoothing uses curve fitting or spline interpolation methods to optimize the path by adjusting key points in the path segment. Based on the spatial distribution of the path segment to be adjusted, multiple candidate paths are set. The path with the lowest cost and best safety compliance is selected from these candidate paths for local replacement. This process effectively avoids the impact of high-corrosion and high-friction areas and reduces the vehicle's driving resistance in complex environments, thereby achieving energy-efficient path optimization. Control points for the locally optimized path segment are selected by minimizing the energy consumption objective function to ensure that the selected path segment achieves optimal energy consumption. During the energy consumption objective function optimization process, factors such as path segment length, friction coefficient, and number of turns are comprehensively considered. Control points are selected to ensure smooth travel while reducing overall energy consumption. After completing control point optimization, the system performs collision detection on the energy-optimized path segment to ensure that the new path segment will not collide with static obstacles or dynamic equipment during actual operation, resulting in a safe and reliable local path segment. Safe local path segments are reconnected with the unmodified segments of the initial optimal path to form an energy-efficient and safe path. During the path splicing process, a path smoothing algorithm is used to smooth the transitions between the connection points, ensuring smooth transitions between path segments without significant speed fluctuations or directional deviations. The cost of the spliced path is also recalculated to ensure optimal energy efficiency and safety for the entire path.

[0032] Step S105: When the AGV is executing the energy-saving and safe path, it monitors the moving obstacles and electrolyte leakage area ahead in real time, calculates the temporary obstacle avoidance direction and speed, and generates a temporary detour path.

[0033] Specifically, a multi-sensor system installed on the AGV (Automated Guided Vehicle) collects real-time workshop environment data. These sensors include a multi-directional distance sensor, a visual recognition system, and a torque sensor. The multi-directional distance sensor detects the position and distance changes of obstacles around the AGV. The visual recognition system identifies the shape, speed, and direction of moving obstacles ahead and can detect the location of electrolyte leaks. The torque sensor is used to detect in real time the torque changes caused by obstacles or ground anomalies during the vehicle's travel. These sensors work together to form a real-time obstacle information collection mechanism. Based on real-time obstacle information, the AGV's current position, and the characteristics of its surroundings, a combined potential field function for the silver electrolysis workshop environment is established. This function is a comprehensive potential field model that combines gravitational and repulsive fields. The gravitational field is used to guide the AGV to its target location, while the repulsive field is used to prevent collisions with obstacles or hazardous areas. The combined potential field function dynamically adjusts the potential field strength in different areas. For example, when an electrolyte leak is detected, the system generates a strong repulsive field in that area, forcing the AGV to move away from the danger zone. Simultaneously, a repulsive field is set in front of the path of moving obstacles to guide the AGV to a safe direction. After establishing the combined potential field function, the negative gradient of the function is calculated and used as the temporary travel direction for the AGV. This negative gradient ensures that the AGV moves in the direction of minimum potential field strength, thereby avoiding obstacles and danger zones, forming a dynamic obstacle avoidance basic direction. After determining the temporary travel direction, a set of candidate speeds is constructed based on the AGV's current state and dynamic constraints. This set of candidate speeds includes different speed combinations that must meet safe driving requirements within the workshop and take into account dynamic constraints such as the AGV's acceleration, deceleration, and turning radius. For each speed combination, a speed score is assigned based on the current driving state and temporary travel direction. The scoring criteria include obstacle avoidance efficiency, path deviation, driving smoothness, and energy optimization. A comprehensive score is given to each speed combination in the speed candidate set, and the speed combination with the highest score is selected as the current real-time obstacle avoidance control instruction. This speed combination can ensure that the transport vehicle maintains smooth operation during the obstacle avoidance process, and can minimize path deviation and energy consumption, thereby achieving efficient obstacle avoidance. When the system detects that the original path is no longer feasible, for example, the path is completely blocked due to the movement of an obstacle, or the path area is no longer safe due to electrolyte leakage, the system re-plans the path in the local area and generates a temporary detour path. Temporary path planning uses a dynamic path planning algorithm to search for a new path that avoids obstacles and meets safety constraints in a limited local area, and seamlessly splices the detour path back to the original energy-saving and safe path, thereby ensuring that the transport vehicle can return to the target driving direction in the shortest time.The generated temporary detour path needs to take into account the current environmental characteristics and ensure path smoothness, steering stability, and smooth connection with the original path to ensure that the AGV transport vehicle does not produce obvious speed fluctuations or direction deviations during its driving process after avoiding obstacles.

[0034] AGV transport tasks are categorized according to the silver electrolysis process flow, dividing them into cathode plate handling and anode plate handling. These tasks correspond to different process requirements and transportation objectives. The cathode plate handling task primarily involves removing the cathode plates from the electrolysis cell and transporting them to the cathode plate storage area, while the anode plate handling task involves transporting the anode plates from the storage area to the electrolysis cell and completing their installation. A leader AGV and multiple follower AGVs are assigned to each task, forming a binary task structure. The leader AGV is responsible for task path planning and execution guidance, while the follower AGVs coordinate movement according to the leader AGV's path instructions, enabling efficient coordinated handling of multiple AGVs within the silver electrolysis process workshop. After establishing this binary task structure, collaborative control is implemented based on a leader-follower control model. This leader-follower architecture ensures stable queue operation by establishing a relative position and velocity relationship between the leader and follower AGVs. The system obtains the real-time position, speed, and steering information of the lead AGV to generate corresponding motion control instructions for each follower AGV. It dynamically adjusts the trajectory and speed of the follower AGVs based on the real-time status of the handling task, thus forming an intra-group collaborative control strategy. During the collaborative control process, time delay compensation and nonlinear fuzzy adaptive control techniques are employed to ensure that the follower AGVs can quickly respond to path adjustments and speed changes, thereby maintaining the stability and accuracy of the queue and avoiding congestion or collisions in the narrow areas of the electrolysis workshop. After completing the intra-group collaborative control strategy design, a task sequence is constructed for all handling tasks in the silver electrolysis process workshop. Tasks are sorted according to their urgency, priority, start and end time windows, and path complexity, and a task sequence matrix is constructed. The task sequence matrix maps all handling tasks into different time windows and arranges them appropriately according to the constraints of each task. Once the task sequence is obtained, it is used as input for task allocation using the Hungarian algorithm. The Hungarian algorithm optimizes task matching by minimizing the total cost of task allocation, assigning each task to the most suitable AGV, and generating an AGV execution plan. The execution plan includes the start and end times of each AGV's mission, the path planning results, and the expected completion time, ensuring task coordination between multiple AGVs. Spatiotemporal conflict detection is performed on the execution plans of each AGV to ensure that different AGVs do not cross paths or have time conflicts during mission execution. Spatiotemporal conflict detection includes path conflict detection, task time overlap detection, and congestion area judgment. Based on the AGV transporter's path and task time window, potential conflicts are detected. When conflicts are found, pre-processing is performed to eliminate conflicts by adjusting the path or task time to obtain a conflict-free execution plan. By introducing dynamic constraints and a safety distance protection mechanism, it is ensured that the paths between different AGVs do not cross at critical time points, ensuring the smooth execution of the transport task.An event-triggered task reallocation mechanism is designed based on a conflict-free execution plan to dynamically respond to unexpected situations during task execution. When the system detects an abnormal event such as a task anomaly, AGV failure, or path obstruction, it automatically triggers a local replanning mechanism. By reanalyzing the current task status and path data, it generates a new path planning solution within the local area and reallocates task resources to form a global collaborative path planning solution.

[0035] In an embodiment of the present application, by collecting environmental characteristics of the silver electrolysis process workshop and constructing a refined electronic map, the electrolytic cell area, corrosive gas area, and equipment-intensive area are accurately distinguished, providing detailed environmental information for the AGV transporter, making the path planning more in line with the actual environmental characteristics of the silver electrolysis workshop. A comprehensive scoring mechanism is used to calculate the initial optimal path, taking into account path length, exposure to the corrosive environment, and equipment collision risk. This reduces the AGV transporter's residence time in high-corrosion areas, extends equipment life, and reduces equipment maintenance costs. The actual driving resistance of the path is dynamically detected and correction parameters are recorded to achieve real-time adaptation to changes in the ground conditions of the silver electrolysis workshop, making the path planning scheme more in line with the complex and changing actual environment of the silver electrolysis workshop. Local adjustments are made based on the path correction parameters to bypass high-corrosion areas and high-friction coefficient areas, reduce energy consumption, and improve the operating efficiency and safety of the AGV transporter in the silver electrolysis workshop. Real-time monitoring of moving obstacles and electrolyte leakage areas ahead, calculation of temporary obstacle avoidance direction and speed, enables the AGV transporter to respond to emergencies in the silver electrolysis workshop in a timely manner, and ensure equipment safety and production continuity. Through the binary task structure and leader-follower architecture, multiple AGV transporters can work together, optimize the transportation process of cathode and anode plates, reduce the empty load rate of AGV transporters, and improve the overall logistics efficiency of the silver electrolysis workshop.

[0036] In a specific embodiment, the process of executing step S101 may specifically include the following steps:

[0037] Based on the physical layout of the silver electrolysis process workshop, we collected information on the location of static obstacles, equipment layout, process pipeline locations, and work area distribution to obtain the original workshop environmental data.

[0038] Perform rasterization processing on the original workshop environment data to obtain the workshop grid model;

[0039] Each cell in the workshop grid model is attribute-labeled to obtain labeled grid data. The attribute labels include free areas, static obstacles, electrolytic cell areas, cathode plate storage areas, anode plate storage areas, short-circuit plate storage areas, channel areas, and areas with high corrosive gas concentrations.

[0040] The marked raster data is converted into a workshop topology diagram, and a weight value is assigned to each edge in the workshop topology diagram according to the actual distance, the corrosiveness coefficient and the friction coefficient of the area where it is located, to obtain a weighted edge set;

[0041] All workstations are marked based on the weighted edge set to obtain an electronic map of the silver electrolysis process workshop. The workstations include the electrolytic cell plate removal station, the cathode plate storage station, and the anode plate storage station.

[0042] Specifically, environmental information is collected from the silver electrolysis workshop. High-precision LiDAR, a visual recognition system, and industrial cameras are deployed at key locations within the workshop, combining panoramic scanning and 3D modeling technologies to collect spatial layout data. The LiDAR scans the entire workshop environment with millimeter-level accuracy, identifying the locations of static obstacles such as walls, equipment, fixtures, and process piping. The visual recognition system uses image recognition technology to confirm the layout of the workshop equipment, including the boundaries and relative positions of key work areas such as the electrolytic cell, cathode plate storage area, anode plate storage area, and short-circuit board storage area. Simultaneously, the layout of the process piping is identified through a combination of laser scanning and visual recognition, ensuring that the spatial orientation and connectivity of the process piping are captured. Multi-sensor fusion technology is used to collect data on the distribution of work areas within the workshop, including the location, boundaries, and environmental characteristics of different functional areas. This data undergoes preliminary processing to form the raw workshop environmental data. The raw workshop environmental data was rasterized, dividing the entire silver electrolysis process workshop into a regular two-dimensional grid model with a grid cell size of 10 cm x 10 cm to ensure sufficient spatial resolution for path planning. The core of the rasterization process is to convert the continuous workshop physical space into a discrete two-dimensional grid model, with each cell corresponding to an actual spatial area within the workshop. This transforms complex environmental information into structured data that can be processed by computers. The attributes of each cell are determined by the environmental data and accomplished through attribute labeling. Attribute labeling is a key step in classifying the grid model according to the characteristics of different workshop areas. Based on the actual environmental information, each cell is assigned a different attribute label, including labels such as free area, static obstacle, electrolytic cell area, cathode plate storage area, anode plate storage area, short-circuit plate storage area, channel area, and area with high corrosive gas concentration. The free area indicates the area where the AGV transporter can pass normally. The static obstacle area is marked as an inaccessible area. The electrolytic cell area needs to be marked separately because it involves important process operations. The cathode and anode plate storage areas are key locations for the transporter to complete the task of taking and placing the plates. The path planning of these areas requires special optimization. The short-circuit plate storage area is an area that requires attention in path planning, and the channel area is the main driving path of the AGV transporter, and the smoothness of the path needs to be ensured. In particular, for areas with high concentrations of corrosive gases, the system will make special marks. In the subsequent path planning process, the transporter should try to avoid such areas to extend the service life of the equipment and ensure safe operation. The marked raster data is converted into a workshop topology structure diagram. By mapping the passable area nodes in the raster data to the vertex set in the topology diagram, and mapping the passage paths between adjacent nodes to the edge set, a weighted directed graph is formed.During the topology map generation process, each edge is precisely weighted. Weights are determined based on the actual distance of the path segment, the corrosivity of the region, and the friction coefficient. The actual distance of a path segment is determined by the Euclidean distance between adjacent nodes. The corrosivity is assigned based on the corrosivity level of the region along the path. Highly corrosive areas are weighted higher to prioritize avoidance during path planning. The friction coefficient is assigned based on the surface material and the characteristics of the path segment. For example, slippery areas or those with high friction coefficients receive higher weights, guiding the transporter to choose the path with the lowest energy consumption. This weighting mechanism dynamically adjusts path weights during subsequent path searches, selecting the optimal path based on the characteristics of each region. Based on this weighted edge set, all workstations are marked to form an electronic map of the silver electrolysis process plant. Workstations include electrolytic cell plate removal stations, cathode plate storage stations, and anode plate storage stations. The electrolytic cell plate retrieval station is the starting point for the AGV to remove the cathode or anode plates from the electrolytic cell. These stations must be precisely identified to ensure the AGV can accurately perform the plate retrieval task. The cathode plate storage station is the target location for the AGV to transport the cathode plates and is also clearly marked on the electronic map. The anode plate storage station is a key station for storing and transporting anode plates and is prioritized in path planning. By mapping these workstations to the vertex set of the topological structure graph, workstation nodes are formed, ensuring that the starting and ending locations can be quickly identified according to the task type during path planning.

[0043] In a specific embodiment, the process of executing step S102 may specifically include the following steps:

[0044] According to the real-time position and task requirements of the AGV in the silver electrolysis process workshop, the starting position and target workstation of the AGV are determined, and the coordinates of the starting and ending points of the path planning are obtained;

[0045] Based on the coordinates of the starting and ending points, a priority queue is constructed on the electronic map of the silver electrolysis process workshop, and the starting position node is added to the queue and the path cost value is initialized to obtain the initial state of the path search;

[0046] Select the node with the smallest replacement value from the priority queue for expansion, calculate the path cost of the adjacent nodes, and obtain the set of candidate path nodes;

[0047] Different travel cost coefficients are assigned to the nodes in the candidate node set according to the type of area they are located in, and a turning penalty factor is introduced to adjust the path cost to obtain the corrected node cost value;

[0048] The safety of each candidate path is checked according to the corrected node cost value to obtain a set of paths that meet the safety constraints. A comprehensive score is then performed based on the set of paths that meet the safety constraints, and the path with the highest score is selected as the initial optimal path.

[0049] Specifically, the AGV positioning system, combined with the task management module, obtains the current position of the AGV and determines its target location based on the current task requirements. The positioning system utilizes a multi-sensor fusion positioning method combining lidar, a visual recognition system, and an inertial measurement unit (IMU). This system determines the real-time coordinates of the AGV by matching it with the electronic map of the silver electrolysis process workshop. Simultaneously, the task management module automatically assigns a task target based on the current task type (e.g., transporting cathode or anode plates), using the coordinates of the target workstation as the endpoint for path planning. The coordinates of the starting and target positions are parsed by the system and used as input to the path planning algorithm, providing the starting and ending coordinates for subsequent path search. Based on the starting and ending coordinates, a priority queue is constructed on the electronic map of the silver electrolysis process workshop. The priority queue is a data structure used to store nodes to be processed during the path search process. Its core function is to sort nodes to be expanded according to their path cost, ensuring that nodes with the lowest cost are processed first. The starting node is added to the priority queue, and its path cost is initialized to zero, indicating that the cost from the starting point to itself is zero, thus achieving the initial state of the path search. During the path search process, a priority queue automatically sorts pending nodes by path cost, ensuring that nodes with the lowest cost are prioritized for each expansion, thereby achieving optimal path search. During the path search process, nodes with the lowest cost are selected from the priority queue for expansion. Specifically, the node with the lowest current path cost is chosen as the expansion node, and the path costs of its neighboring nodes are calculated to generate a set of candidate nodes. Candidate nodes are the neighbors of the current node, representing all possible path points reachable via a single-step expansion. The system calculates the costs of candidate nodes based on the current node's location, the locations of its neighboring nodes, and path characteristics, and adds them to the candidate node set. For each candidate node, its path cost is evaluated to ensure that more optimal paths are prioritized during the path search, thereby continuously approaching the target location. Nodes in the candidate node set are assigned different travel cost coefficients based on the area they are located in. The travel cost coefficient reflects the difficulty and energy consumption of navigating different areas. For example, the travel cost coefficient is lower in access areas, while it is higher in electrolytic cell areas or cathode plate storage areas to prevent frequent travel by trucks in these complex areas. A turn penalty factor is introduced to adjust the path cost. When a turn occurs during path expansion, the turn penalty factor is calculated based on the turn angle. Larger turns increase the penalty factor, ensuring that the path planning algorithm chooses smoother paths and avoids unnecessary sharp turns. After path cost correction, a revised node cost value is obtained, reflecting the actual cost of the candidate path. Each candidate path is then safety-tested based on this revised node cost value to ensure that it meets the safety constraints of the silver electrolysis process workshop.Safety checks include multiple factors, such as the minimum safe distance between the detection path and static obstacles, whether the detection path passes through areas with high concentrations of corrosive gases, and whether there are other potential safety risk areas. Candidate paths are screened based on safety constraints, and paths that meet safety requirements are retained to obtain a set of paths that meet safety constraints. A comprehensive score is then assigned to the set of paths that meet safety constraints, based on multiple criteria including path length, path cost, driving smoothness, obstacle avoidance success rate, and energy consumption optimization. A comprehensive score is calculated for each candidate path by taking a weighted sum of the scores from different dimensions, and the path with the highest score is selected as the initial optimal path.

[0050] In a specific embodiment, the process of executing step S103 may specifically include the following steps:

[0051] The sensor group installed on the AGV transport vehicle collects real-time driving data to obtain the original data of the AGV transport vehicle's driving status. The sensor group includes distance sensor data, visual recognition system data, and torque sensor data;

[0052] Segment the original data of the AGV transporter's driving status according to the path segments to obtain segmented driving parameters;

[0053] The momentum residual value of each path segment is calculated according to the segmented driving parameters to obtain the path segment momentum residual data, where the momentum residual value is equal to the absolute value of the difference between the mass of the AGV transporter multiplied by the average speed and the force-time product;

[0054] Compare the momentum residual data of the path segment with a preset momentum residual threshold. If the momentum residual value exceeds the momentum residual threshold, mark the path segment as an abnormal path segment to obtain an abnormal path segment set.

[0055] The adaptive learning rate is calculated according to the momentum residual value of each path segment in the abnormal path segment set to obtain the parameter update coefficient, and the ground friction coefficient of each path segment in the abnormal path segment set is corrected and calculated based on the parameter update coefficient to obtain the path correction parameter.

[0056] Specifically, the AGV is integrated with multiple high-precision sensors, including distance sensors, a visual recognition system, and torque sensors. These sensors collect real-time workshop environment information, path travel data, and dynamic status data while the vehicle is performing its transport tasks. The distance sensors detect the location of obstacles around the vehicle, path boundaries, and changes in distance to dynamic objects. This data helps the system determine the spatial traffic conditions of the current path segment. The visual recognition system, through real-time image acquisition and analysis of the workshop environment, identifies the locations of obstacles, electrolytic cells, cathode plates, anode plates, and pipelines in the forward area and determines whether any anomalies have occurred in the current path segment. Simultaneously, the torque sensor measures the friction and resistance between the vehicle and the ground, as well as changes in torque generated during acceleration and deceleration, to obtain data on the vehicle's force status during travel. All of this data is integrated by a multi-sensor data fusion system and, coordinated by a time synchronization module, forms raw data on the AGV's driving status. The system segments the raw data of the AGV's driving status according to path segments, precisely classifying the vehicle's travel data along different path segments. Each path segment corresponds to an independent spatial region or path interval that the vehicle traverses within the silver electrolysis process workshop. Segmentation is a key step in path analysis. The path segment identification module divides the driving data into different regions. Each path segment includes parameters such as travel time, average speed, friction change, and torque change. This segmented data provides input for the subsequent momentum residual calculation. Based on the path segment's topological information and start and end point coordinates, the system splits the raw driving data into multiple independent path segments and stores the parameters of each path segment in a path segment data set for subsequent calculations. After segmentation, the momentum residual is calculated for each path segment based on the segmented driving parameters. The momentum residual is a key indicator of the degree of abnormal force state during the vehicle's travel. The calculation method is to multiply the AGV's mass by the average speed of the path segment, compare it with the force-time product of the path segment, and calculate the absolute value of the difference between the two. The momentum residual value reflects the deviation between the actual driving resistance and the theoretical resistance of the transport vehicle on the path segment. If the momentum residual value is large, it means that there is a large resistance change on the path segment. The momentum residual data of the path segment is compared with the preset momentum residual threshold. The momentum residual threshold is set according to the environmental characteristics of the silver electrolysis workshop, the dynamic characteristics of the transport vehicle, and the momentum change range under normal operating conditions. When the system detects that the momentum residual value of a path segment exceeds the preset threshold, the path segment is marked as an abnormal path segment and the abnormal path segment is recorded in the abnormal path segment set. The cause of the abnormality is analyzed based on the abnormal path segment set, and the friction coefficient of the path segment is dynamically adjusted. Abnormal path segments are concentrated in high-friction areas, path blockage points or corrosive areas. These areas are likely to cause increased energy consumption and decreased driving stability during the transport vehicle's driving process.An adaptive learning rate is calculated based on the momentum residual value of each path segment in the abnormal path segment set. The adaptive learning rate is a dynamically adjusted parameter that adaptively adjusts the friction coefficient correction step size for each path segment based on the magnitude of the momentum residual. The learning rate calculation follows an exponential decay mechanism: when the momentum residual value is large, the learning rate is higher, thereby accelerating the correction of the path segment's friction coefficient. When the momentum residual value is small, the learning rate is gradually reduced to ensure the stability of the path correction. By adjusting the adaptive learning rate, the friction coefficient correction speed of each path segment is kept consistent with the change in the momentum residual, thereby improving the accuracy and efficiency of path optimization. Based on the parameter update coefficient, the ground friction coefficient of each path segment in the abnormal path segment set is corrected and calculated. By dynamically updating the friction coefficient, the impact of abnormal resistance in the path segment is eliminated, thereby optimizing the path planning effect. The friction coefficient is adaptively updated based on the magnitude of the momentum residual of the abnormal path segment and the driving parameters of the path segment. Through repeated iterative corrections, the friction coefficient of the path segment is stabilized and the path correction parameters are formed. These correction parameters are fed back into the electronic map of the silver electrolysis workshop to update the path weights of the corresponding areas, thereby guiding the AGV transporter to choose a better path and avoid high-friction areas or areas with abnormal resistance during the subsequent path planning process.

[0057] In a specific embodiment, the process of executing step S104 may specifically include the following steps:

[0058] Feeding back the path correction parameters to the electronic map of the silver electrolysis process workshop, correcting the edge weights of the corresponding areas, and obtaining updated electronic map weight data;

[0059] Identify a set of path segments to be adjusted based on the updated electronic map weight data and the initial optimal path;

[0060] Each path segment in the set of path segments to be adjusted is smoothed to bypass high corrosion areas and high friction coefficient areas to obtain a local optimized path segment;

[0061] By minimizing the energy consumption objective function, the control points of the local optimized path segment are selected to obtain the local path with the optimal energy consumption. The local path with the optimal energy consumption is then subjected to collision detection verification to obtain a safe local path segment.

[0062] The safe local path segments are reconnected with the unmodified path segments in the initial optimal path to obtain an energy-saving safe path.

[0063] Specifically, the path correction parameters identified by the momentum residual detection module are fed back to the electronic map system of the silver electrolysis process workshop, and the weighted edge set in the electronic map is dynamically updated. The path correction parameters include the friction coefficient, regional corrosion coefficient, and path resistance anomaly information for each path segment. These parameters directly affect the calculation of edge weights during path planning. Therefore, the weights of the corresponding path segments are corrected based on the path correction parameters. The core of the weight correction is to appropriately increase the weight values of high-friction areas, high-corrosion areas, and areas with abnormal path resistance, thereby guiding the AGV transport vehicle to preferentially bypass these unfavorable areas during the path planning process. The updated electronic map weight data can reflect the latest workshop environmental status. Based on the updated electronic map weight data and the initial optimal path, the set of path segments to be adjusted is identified. The initial optimal path is a path generated based on the original electronic map. Due to changes in the path correction parameters, some path segments are no longer the path with the highest energy consumption or safety, so these path segments are re-identified. The path segment analysis module calculates updated path costs and identifies path segments with abnormal costs as those to be adjusted. These path segments are located in areas of high corrosion, varying friction coefficients, or abnormal resistance. These areas significantly increase the path cost, directly impacting the AGV's operating efficiency and safety. Each path segment in the set of path segments to be adjusted is then smoothed to ensure smoothness and safety as the AGV navigates through high-corrosion and high-friction areas. Path smoothing utilizes B-spline or Bezier curves, optimizing the path curve by adjusting control points to achieve smooth transitions. The core objectives of path smoothing are to ensure path continuity and steering smoothness, preventing sharp turns and path fluctuations in high-corrosion or high-friction areas, reducing energy consumption, and improving the AGV's stability during path execution. During the path smoothing process, multiple candidate paths are generated based on the spatial distribution of the path segments. The candidate path evaluation module comprehensively compares the smoothness, steering angle, and energy consumption of the different candidate paths. The optimal path smoothing solution is selected, resulting in a locally optimized path segment. By minimizing the energy consumption objective function, the control points of the local optimized path segment are selected to ensure that the path reaches the optimal state in terms of energy consumption. The energy consumption minimization objective function is the core constraint of path optimization. Its optimization goal is to minimize the driving energy consumption of the transporter by adjusting the control point positions of the path segment, reducing the path length, reducing friction resistance, and reducing turning penalties. During the optimization process, the system iteratively adjusts the control points of the candidate path segments, and combines the path cost evaluation model to calculate the path energy consumption under different control point configurations, and selects the local path segment with the optimal energy consumption. The local path with the optimal energy consumption is collided and verified to ensure that the path segment will not collide with obstacles or equipment in the actual environment.Collision detection is implemented using a time-based dynamic window method combined with a static obstacle detection algorithm. During the detection process, dynamic simulation is performed based on the transporter's motion model, the control points of the path segments, and the positions of surrounding obstacles to ensure that the transporter avoids collisions with static or dynamic obstacles while traveling along the optimized path segment. The system also checks the minimum safe distance of the path segments to ensure that a safe distance is always maintained between the transporter and obstacles. When a potential collision risk is detected in a path segment, the system automatically makes local adjustments to the path segment and re-runs collision detection until a safe local path segment is obtained. The safe local path segment is reconnected with the unmodified path segments from the initial optimal path to form an energy-efficient and safe path. The path segment reconnection module seamlessly splices the path segments based on their spatial position and time sequence, ensuring smooth transitions between the connection points and eliminating sudden speed changes or directional deviations caused by the splicing process. The total path length, travel time, and estimated energy consumption are recalculated based on the spliced path segments to ensure the overall performance of the spliced energy-efficient and safe path is optimized.

[0064] In a specific embodiment, the process of executing step S105 may specifically include the following steps:

[0065] The multi-directional distance sensor, visual recognition system and torque sensor on the AGV transporter collect real-time environmental data, identify the position and speed of moving obstacles in front and the location of electrolyte leakage areas, and obtain real-time obstacle information;

[0066] Based on real-time obstacle information and the current position of the AGV, a combined potential field function is established in the silver electrolysis workshop environment.

[0067] Calculate the negative gradient value of the resultant potential field function, use the negative gradient direction as the temporary moving direction of the AGV transporter, and obtain the basic obstacle avoidance direction;

[0068] Based on the current state and dynamic constraints of the AGV, a speed candidate set is constructed, and speed score data is calculated for each speed combination in the speed candidate set;

[0069] The speed combination with the highest score is selected based on the speed score data to obtain real-time obstacle avoidance control instructions. When it is detected that the original path is no longer feasible, the path is replanned in the local area to generate a temporary detour path.

[0070] Specifically, a multi-sensor data fusion system uses multi-directional distance sensors integrated into the AGV to monitor the distance, shape, and position changes of obstacles in front of and around the vehicle in real time. Simultaneously, a visual recognition system uses image processing algorithms to classify and track obstacles, identifying the position, speed, and direction of moving obstacles. A torque sensor detects torque fluctuations caused by road surface changes or abnormal resistance during the vehicle's travel, assisting in determining the location of electrolyte leaks. Through the synergistic effect of multi-sensor data, real-time obstacle information is acquired, including the obstacle's spatial coordinates, movement speed, trajectory, and the extent and location of the leak area. A combined potential field function is constructed based on this real-time obstacle information, the AGV's current position, and the characteristics of the workshop environment. This combined potential field function consists of a gravitational field and a repulsive field. The gravitational field, with the target location as the source of attraction, generates an attractive force that guides the vehicle toward the target, while the repulsive field generates a repulsive force around obstacles and dangerous areas, forcing the vehicle away from these areas. A stronger repulsive field is applied to electrolyte leak areas to ensure that the vehicle prioritizes avoiding these areas during route adjustment. The combined potential field is constructed by fully considering the distance between the AGV and the obstacle, the obstacle's speed, and the complexity of the path. The potential field strength is dynamically adjusted based on the characteristics of different areas within the workshop, enabling the system to generate a reasonable obstacle avoidance path in real time in complex environments. As the AGV's position changes, the potential field function is continuously updated to ensure real-time and accurate obstacle avoidance decisions. After constructing the combined potential field function, the negative gradient of the function is calculated, and the direction of the negative gradient is used as the temporary travel direction for the AGV to generate the basic obstacle avoidance direction. The negative gradient direction is the direction with the fastest descent along the potential field function. This means that in real-world environments, this direction effectively avoids obstacles and dangerous areas while maintaining guidance to the target location. By calculating the negative gradient in real time, the AGV's travel direction is rapidly adjusted, ensuring that it maintains a stable and reasonable path during obstacle avoidance. Even in the face of dynamic obstacles or sudden changes in the path, the calculated negative gradient direction provides rapid guidance, helping the AGV make obstacle avoidance decisions in the shortest possible time. Based on the determined basic obstacle avoidance direction, a set of candidate speeds is constructed, taking into account the AGV's current state and dynamic constraints. The truck's status includes information such as its current position, speed, acceleration, and steering angle. Dynamic constraints involve parameters such as the vehicle's maximum speed, acceleration limit, and minimum turning radius. Based on these constraints, the system generates multiple possible speed combinations, each consisting of linear and angular velocities, to form a set of speed candidates. By considering the impact of different speed combinations on the truck's driving stability and obstacle avoidance capabilities, the system provides the truck with greater freedom of movement while maintaining safety, enabling flexible adjustments during obstacle avoidance.After generating a set of candidate speeds, a speed score is calculated for each speed combination in the set. The scoring criteria include obstacle avoidance efficiency, path deviation, ride smoothness, and energy optimization. Obstacle avoidance efficiency primarily measures the success rate of the truck in avoiding obstacles after selecting a specific speed combination; path deviation assesses the extent to which the truck deviates from the planned path during obstacle avoidance; ride smoothness considers the truck's smoothness during cornering, acceleration, and deceleration; and energy optimization focuses on the truck's energy consumption under different speed combinations. The system then weights and sums the scores for all speed combinations to calculate a comprehensive score for each speed combination. Based on the speed score data, the highest-scoring speed combination is selected as the current real-time obstacle avoidance control command, which includes specific linear and angular velocity control values. This command is then sent to the truck's control system for execution. During the execution of the obstacle avoidance control command, the system monitors the position of obstacles and the truck's driving status in real time to ensure the effectiveness and safety of the obstacle avoidance path. Once the truck avoids an obstacle and returns to its normal path, the system automatically terminates obstacle avoidance mode and returns to the original path planning command. If the system detects that the original path is no longer feasible due to obstacles or sudden changes in the environment, the system triggers the local path replanning module, replans the path in the local area, and generates a temporary detour path. The goal of local path replanning is to find a safe path for the transport vehicle that avoids obstacles in the shortest time, while minimizing the path length and driving time. Based on the current electronic map and real-time obstacle information, the system reconstructs the path search space and uses a dynamic programming algorithm or an A* algorithm for path search. In order to ensure the rationality of the temporary detour path, the path is smoothed to avoid sharp turns and path discontinuities. In addition, collision detection and safety verification of the path are performed to ensure that the generated temporary path does not conflict with other obstacles.

[0071] In a specific embodiment, the path planning method for an AGV transport vehicle in a silver electrolysis process workshop further includes the following steps:

[0072] According to the silver electrolysis process, the AGV handling task is divided into the cathode plate handling task and the anode plate handling task. A leading AGV and multiple following AGVs are assigned to the cathode plate handling task and the anode plate handling task respectively, thus obtaining a binary task structure.

[0073] A leader-follower architecture control model is constructed based on the binary task structure, and a collaborative control strategy within the group is generated based on the leader-follower architecture control model.

[0074] A task sequence is constructed for all handling tasks in the silver electrolysis process workshop, and the task sequence is input into the Hungarian algorithm for task allocation to obtain the AGV execution plan;

[0075] Perform spatiotemporal conflict detection on each AGV execution plan to obtain a conflict-free execution plan;

[0076] An event-triggered task reallocation mechanism is designed based on a conflict-free execution scheme. When task anomalies, AGV failures, or path blockages are detected, local replanning is performed to obtain a global collaborative path planning scheme.

[0077] Specifically, handling tasks are divided according to the production task characteristics of the silver electrolysis process. Cathode plate handling tasks primarily involve removing cathode plates from the electrolytic cell and transporting them to the cathode plate storage area, while anode plate handling tasks involve delivering new anode plates to the electrolytic cell and replacing them. These two tasks have different process requirements and handling paths, so a bipartite task structure is constructed based on task type. In this bipartite task structure, each task is assigned a leader AGV and multiple follower AGVs to jointly complete the task. The leader AGV is responsible for path planning, target recognition, and task status updates, while the follower AGVs perform motion control based on the leader AGV's path and speed commands, achieving efficient coordination of multi-vehicle platoons. After constructing the bipartite task structure, multi-AGV coordinated control is implemented based on a leader-follower architecture control model. The leader-follower architecture ensures stable and consistent platoon operation by establishing relative position information, speed matching, and path synchronization between the leader and follower AGVs. The leader AGV obtains its own position information, speed status, and task progress in real time and transmits this data as control signals to the follower AGVs. The following AGVs adaptively adjust their distance, speed, and path angle differences with the leading AGV, maintaining tight and stable coordination within the entire fleet. The system uses a fleet control algorithm to regulate the distance between the following AGV and the leading AGV, ensuring fleet separation or collisions during turns, acceleration, and deceleration. A time delay compensation mechanism is employed to address control lag caused by communication delays, thereby generating an intra-group collaborative control strategy to ensure stable and efficient multi-vehicle task execution. A task sequence is constructed for all handling tasks in the silver electrolysis process workshop. This task sequence is based on multiple factors, including task priority, task completion time window, path length, energy consumption estimate, and equipment status. The system ranks all handling tasks based on these factors and constructs a task sequence matrix. This matrix maps each task's priority, execution time, and resource requirements, providing a reference for task allocation. The task sequence is then fed into the Hungarian algorithm for task allocation. The Hungarian algorithm is an optimization algorithm designed to achieve optimal matching in complex multi-task, multi-resource environments. Using the task allocation matrix as input, the Hungarian algorithm solves the problem based on task execution costs, time constraints, and resource load, obtaining the optimal task allocation and generating an AGV execution plan. This plan includes each AGV's task starting point, target location, execution sequence, and estimated completion time, ensuring that all tasks are completed on time and that workshop resources are fully utilized. Spatiotemporal conflict detection is performed on each AGV's execution plan to ensure that multi-vehicle collaborative operations avoid path conflicts or resource preemption due to time overlap. This spatiotemporal conflict detection includes multiple aspects, including path conflict detection, task time overlap detection, and dynamic obstacle prediction.The system cross-checks each AGV's execution path with the paths of other AGVs and calculates the time difference between intersecting path segments. If the time difference falls below a set safety threshold, a conflict is considered. Furthermore, the start and end times of each task are checked for overlap to ensure that the same area or equipment is not occupied by different AGVs at the same time, resulting in a conflict-free execution plan. After spatial and temporal conflict detection, path adjustments or task times are fine-tuned based on the detection results to ensure that all AGVs maintain spatial and temporal independence during execution, improving the overall coordination of the handling task. Based on the conflict-free execution plan, an event-triggered task reassignment mechanism is designed to address unexpected situations during task execution, such as task anomalies, AGV failures, or path obstructions. The event-triggered mechanism continuously monitors the operating status of the AGVs through a real-time monitoring module. When the system detects a task anomaly or an AGV failure, it automatically triggers the task reassignment mechanism. The task reassignment mechanism reassesses the current task status and remaining resources, analyzes the priority of the affected tasks, uncompleted paths, and the idleness of available AGVs, and then re-invokes the Hungarian algorithm to perform local task reassignment based on the new task allocation matrix. During the local reallocation process, abnormal tasks are preferentially assigned to the nearest backup AGV to ensure that the task can be resumed in the shortest time possible. When a path blockage is detected, the system activates the local path replanning module, dynamically searches the affected path segments, and generates a new temporary detour path to ensure that the AGV can bypass the obstacle and continue to complete the task. Local path replanning uses the A* algorithm or Dijkstra algorithm for path search, combined with a dynamic obstacle avoidance mechanism to ensure that the new path strikes a balance between obstacle avoidance and driving efficiency. After completing task reallocation and path replanning, a global collaborative path planning solution is generated. The solution includes the reallocation results of all tasks, as well as the new execution plan and conflict detection results after the path adjustment. The global collaborative path planning solution ensures efficient collaborative operation of multiple AGVs in complex workshop environments by reintegrating all task status and path information. Even in the event of task anomalies or path blockages, emergency adjustments can be made quickly to maintain the consistency and reliability of task execution.

[0078] The above describes the path planning method of the AGV transport vehicle used in the silver electrolysis process workshop in the embodiment of the present application. The following describes the path planning system of the AGV transport vehicle used in the silver electrolysis process workshop in the embodiment of the present application. Figure 2 In one embodiment of the present application, a path planning system for an AGV transport vehicle in a silver electrolysis process workshop includes:

[0079] A construction module 201 is used to collect environmental characteristics of the silver electrolysis process workshop and construct an electronic map to obtain an electronic map of the silver electrolysis process workshop;

[0080] A calculation module 202 is used to calculate the initial optimal path of the AGV transport vehicle from the current position to the target workstation based on the electronic map of the silver electrolysis process workshop;

[0081] The detection module 203 is used to detect the actual driving resistance during the execution of the initial optimal path. When the abnormal resistance value of a path segment exceeds the momentum residual threshold, the actual friction coefficient and obstacle conditions of the path segment are recorded to obtain the path correction parameters;

[0082] A local adjustment module 204 is configured to locally adjust the initial optimal path by bypassing high corrosion areas and high friction coefficient areas based on the path correction parameters to obtain an energy-saving and safe path;

[0083] The real-time monitoring module 205 is used to monitor the moving obstacles and electrolyte leakage areas ahead in real time while the AGV transport vehicle is executing the energy-saving and safe path, and calculate the temporary obstacle avoidance direction and speed to generate a temporary detour path.

[0084] Through the collaborative efforts of these components, the system collects environmental characteristics of the silver electrolysis workshop and constructs a refined electronic map to accurately distinguish between electrolytic cell areas, corrosive gas areas, and equipment-intensive areas. This provides detailed environmental information to the AGV, enabling path planning more tailored to the actual environment of the silver electrolysis workshop. A comprehensive scoring mechanism is used to calculate the initial optimal path, taking into account path length, exposure to corrosive environments, and equipment collision risk. This reduces the AGV's dwell time in high-corrosion areas, extending equipment life and reducing maintenance costs. Dynamic detection of actual path resistance and recording of correction parameters enable real-time adaptation to changing floor conditions within the silver electrolysis workshop, ensuring path planning more closely aligned with the complex and changing environment. Local adjustments based on path correction parameters are made to bypass high-corrosion and high-friction areas, reducing energy consumption and improving the efficiency and safety of AGVs within the workshop. Real-time monitoring of moving obstacles and electrolyte leaks ahead allows for calculation of temporary obstacle avoidance directions and speeds, enabling the AGV to respond promptly to emergencies within the workshop, ensuring equipment safety and production continuity. Through the binary task structure and leader-follower architecture, multiple AGV transporters can work together, optimize the transportation process of cathode and anode plates, reduce the empty load rate of AGV transporters, and improve the overall logistics efficiency of the silver electrolysis workshop.

[0085] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described systems, systems and units can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0086] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a path planning device for an AGV transport vehicle used in a silver electrolysis process workshop (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0087] As described above, the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that the technical solutions described in the above embodiments can still be modified, or some of the technical features thereof can be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A path planning method for an AGV transport vehicle in a silver electrolysis process workshop, characterized in that: include: Collect environmental characteristics and construct an electronic map of the silver electrolysis process workshop to obtain an electronic map of the silver electrolysis process workshop; Calculate the initial optimal path of the AGV transport vehicle from the current position to the target workstation according to the electronic map of the silver electrolysis process workshop; detecting the actual driving resistance during the execution of the initial optimal path, and when the abnormal resistance value of a path segment exceeds a momentum residual threshold, recording the actual friction coefficient and obstacle conditions of the path segment to obtain a path correction parameter; According to the path correction parameters, the initial optimal path is locally adjusted by bypassing high corrosion areas and high friction coefficient areas to obtain an energy-saving and safe path; When the AGV transport vehicle executes the energy-saving and safe path, it monitors the moving obstacles and electrolyte leakage areas ahead in real time, calculates the temporary obstacle avoidance direction and speed, and generates a temporary detour path.

2. The path planning method for an AGV transport vehicle in a silver electrolysis process workshop according to claim 1, characterized in that: The environmental characteristics of the silver electrolysis process workshop are collected and an electronic map is constructed to obtain an electronic map of the silver electrolysis process workshop, including: Based on the physical layout of the silver electrolysis process workshop, we collected information on the location of static obstacles, equipment layout, process pipeline locations, and work area distribution to obtain the original workshop environmental data. Performing rasterization processing on the original workshop environment data to obtain a workshop raster model; Performing attribute marking on each cell in the workshop grid model to obtain marked grid data, wherein the attribute markings include free area, static obstacle, electrolytic cell area, cathode plate storage area, anode plate storage area, short-circuit plate storage area, channel area, and area with high corrosive gas concentration; Converting the marked raster data into a workshop topology diagram, and assigning a weight value to each edge in the workshop topology diagram according to the actual distance, the corrosiveness coefficient and the friction coefficient of the area where the edge is located, to obtain a weighted edge set; All workstations are marked based on the weighted edge set to obtain an electronic map of the silver electrolysis process workshop, wherein the workstations include an electrolytic cell plate removal station, a cathode plate storage station, and an anode plate storage station.

3. The path planning method for an AGV transport vehicle in a silver electrolysis process workshop according to claim 1, characterized in that: The calculating of the initial optimal path of the AGV transport vehicle from the current position to the target workstation according to the electronic map of the silver electrolysis process workshop includes: According to the real-time position and task requirements of the AGV in the silver electrolysis process workshop, the starting position and target workstation of the AGV are determined, and the coordinates of the starting and ending points of the path planning are obtained; Building a priority queue on the electronic map of the silver electrolysis process workshop based on the coordinates of the starting and ending points, adding the starting position node to the queue and initializing the path cost value to obtain an initial state of the path search; Selecting a node with the smallest replacement value from the priority queue for expansion, calculating the path costs of adjacent nodes, and obtaining a set of candidate path nodes; Assigning different travel cost coefficients to the nodes in the candidate node set according to the type of area they are located in, and introducing a turning penalty factor to adjust the path cost to obtain a revised node cost value; A safety check is performed on each candidate path according to the corrected node cost value to obtain a set of paths that meet the safety constraints, and a comprehensive score is performed based on the set of paths that meet the safety constraints, and the path with the highest score is selected as the initial optimal path.

4. The path planning method for an AGV transport vehicle in a silver electrolysis process workshop according to claim 1, characterized in that: The actual driving resistance during the execution of the initial optimal path is detected. When the abnormal resistance value of the path segment exceeds the momentum residual threshold, the actual friction coefficient and obstacle conditions of the path segment are recorded to obtain the path correction parameters, including: The sensor group installed on the AGV transport vehicle collects real-time driving data to obtain the original data of the AGV transport vehicle's driving status. The sensor group includes distance sensor data, visual recognition system data, and torque sensor data; Segmenting the original data of the AGV's driving state according to the path segments to obtain segmented driving parameters; Calculating the momentum residual value of each path segment according to the segmented driving parameters to obtain the path segment momentum residual data, wherein the momentum residual value is equal to the absolute value of the difference between the mass of the AGV transport vehicle multiplied by the average speed and the force-time product; Comparing the momentum residual data of the path segment with a preset momentum residual threshold, and if the momentum residual value exceeds the momentum residual threshold, marking the path segment as an abnormal path segment to obtain an abnormal path segment set; An adaptive learning rate is calculated based on the momentum residual value of each path segment in the abnormal path segment set to obtain a parameter update coefficient, and a ground friction coefficient of each path segment in the abnormal path segment set is corrected and calculated based on the parameter update coefficient to obtain a path correction parameter.

5. The path planning method for an AGV transport vehicle in a silver electrolysis process workshop according to claim 1, characterized in that: The method of locally adjusting the initial optimal path by bypassing high corrosion areas and high friction coefficient areas based on the path correction parameters to obtain an energy-saving and safe path includes: Feeding back the path correction parameters to the electronic map of the silver electrolysis process workshop, correcting the edge weights of the corresponding areas, and obtaining updated electronic map weight data; Identifying a set of path segments to be adjusted based on the updated electronic map weight data and the initial optimal path; Smoothing each path segment in the set of path segments to be adjusted, bypassing high corrosion areas and high friction coefficient areas, and obtaining a locally optimized path segment; Selecting control points of the local optimized path segment by minimizing the energy consumption objective function to obtain a local path with optimal energy consumption, and performing collision detection verification on the local path with optimal energy consumption to obtain a safe local path segment; The safe local path segment is reconnected with the unmodified path segment in the initial optimal path to obtain an energy-saving safe path.

6. The path planning method for an AGV transport vehicle in a silver electrolysis process workshop according to claim 1, characterized in that: In the process of the AGV transport vehicle executing the energy-saving safety path, the moving obstacles and electrolyte leakage area in front are monitored in real time, and a temporary obstacle avoidance direction and speed are calculated to generate a temporary detour path, including: The multi-directional distance sensor, visual recognition system and torque sensor on the AGV transport vehicle collect real-time environmental data, identify the position and speed of the moving obstacle in front and the location of the electrolyte leakage area, and obtain real-time obstacle information; Establishing a resultant potential field function in a silver electrolysis workshop environment according to the real-time obstacle information and the current position of the AGV transport vehicle; Calculating the negative gradient value of the resultant potential field function, taking the negative gradient direction as the temporary moving direction of the AGV transport vehicle, and obtaining the basic obstacle avoidance direction; Based on the current state and dynamic constraints of the AGV, a speed candidate set is constructed, and speed score data is calculated for each speed combination in the speed candidate set; The speed combination with the highest score is selected according to the speed score data to obtain a real-time obstacle avoidance control instruction. When it is detected that the original path is no longer feasible, the path is re-planned in the local area to generate a temporary detour path.

7. The path planning method for an AGV transport vehicle in a silver electrolysis process workshop according to claim 1, characterized in that: The path planning method for the AGV transport vehicle in the silver electrolysis process workshop also includes: According to the silver electrolysis process, the AGV handling task is divided into a cathode plate handling task and an anode plate handling task, and a leader AGV and multiple follower AGVs are assigned to the cathode plate handling task and the anode plate handling task respectively, thereby obtaining a binary task structure; Constructing a leader-follower architecture control model based on the binary task structure, and generating an intra-group collaborative control strategy according to the leader-follower architecture control model; Constructing a task sequence for all handling tasks in the silver electrolysis process workshop, and inputting the task sequence into the Hungarian algorithm for task allocation to obtain an AGV execution plan; Perform spatiotemporal conflict detection on each AGV execution plan to obtain a conflict-free execution plan; An event-triggered task reallocation mechanism is designed based on the conflict-free execution scheme. When task anomalies, AGV failures, or path blockages are detected, local replanning is performed to obtain a global collaborative path planning scheme.

8. A path planning system for an AGV transport vehicle in a silver electrolysis process workshop, characterized in that: A method for implementing a path planning system for an AGV transport vehicle used in a silver electrolysis process workshop according to any one of claims 1 to 7, wherein the path planning system for an AGV transport vehicle used in a silver electrolysis process workshop comprises: A construction module is used to collect environmental characteristics of the silver electrolysis process workshop and construct an electronic map to obtain an electronic map of the silver electrolysis process workshop; A calculation module, configured to calculate an initial optimal path for the AGV transport vehicle from a current position to a target workstation based on an electronic map of the silver electrolysis process workshop; a detection module, configured to detect actual driving resistance during execution of the initial optimal path, and when an abnormal resistance value of a path segment exceeds a momentum residual threshold, record the actual friction coefficient and obstacle conditions of the path segment, and obtain a path correction parameter; A local adjustment module is used to locally adjust the initial optimal path by bypassing high corrosion areas and high friction coefficient areas according to the path correction parameters to obtain an energy-saving and safe path; The real-time monitoring module is used to monitor the moving obstacles and electrolyte leakage areas ahead in real time while the AGV transport vehicle is executing the energy-saving and safe path, and calculate the temporary obstacle avoidance direction and speed to generate a temporary detour path.

Images