Integrated algorithm for distribution and scheduling of AGV (Automatic Guided Vehicle) carrying tasks

Abstract

The invention discloses a comprehensive algorithm for distribution and scheduling of AGV (Automatic Guided Vehicle) carrying tasks. S1: an SCADA system; equipment communication, data acquisition and storage, protocol analysis and logic processing; s11, the communication protocol, the address and the communication rhythm of the equipment are configured through the system, data acquisition data are stored in ES (ElasticSearch), the data acquisition data are stored for one year and backed up regularly, and the system achieves service logic processing through an LUA script. The carrying efficiency of the AGV is integrally improved, and the execution time of carrying a single task by the AGV is shortened by 10%; the production capacity of a workshop is improved, and the carrying rate exceeds 96% under the condition of no interference; the workshop logistics cost is reduced, and the overall number of AGVs is reduced by 10%; and the first pass yield of material carrying is improved by 15%. The deep fusion data analysis capability enables the whole system to predict demands more accurately, avoid conflicts, optimize the overall production efficiency, optimize the production process, reduce waiting and resource waste, and improve the flexibility of path planning and the reliability of task completion.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent logistics of photovoltaic automated workshops, and in particular to a comprehensive algorithm for dispatching and scheduling AGV handling tasks. Background Art

[0002] The photovoltaic industry is developing rapidly around the world, and the market size continues to expand. As the photovoltaic industry continues to grow, the demand for automated production is becoming increasingly urgent.

[0003] Competition in the solar photovoltaic industry is fierce, and reducing overall production costs has become an important requirement for photovoltaic production. In order to improve production efficiency, reduce costs and ensure product quality, photovoltaic workshops need to further improve the efficiency of AGV material handling and reduce handling costs. This requires the deep integration of the FMS system and the RCS system. According to the development of the market and business, robot equipment from different manufacturers is scheduled on the same site. It is necessary to solve the problem that customers need different scheduling systems for different manufacturers and equipment for robots and robots of different models from different manufacturers. All robots can be scheduled with one scheduling system, thereby achieving supply chain optimization and upgrading of the industry, reducing customer investment costs, and improving the overall automation and intelligent technical capabilities of the industry.

[0004] Currently, no effective solution has been proposed for the problems in the related technologies. Summary of the invention

[0005] In view of the problems in the related art, the present invention proposes a comprehensive algorithm for dispatching and scheduling AGV handling tasks to overcome the above technical problems existing in the existing related art.

[0006] To this end, the specific technical solution adopted by the present invention is as follows: A comprehensive algorithm for dispatching and scheduling AGV handling tasks includes the following steps: S1: SCADA system; equipment communication, data acquisition and storage, protocol analysis, logic processing; S11: The communication protocol, address, and communication rhythm of the equipment are configured through the system. The data is stored in ES (ElasticSearch). The data is kept for one year and backed up regularly. The system implements business logic processing through LUA scripts. S12: Event triggering: set the event triggering conditions through LUA scripts, and push the triggering events to the consumer end through MQ queues for processing, realizing distributed business logic processing, reducing coupling, and being flexible without affecting mainstream business. Such as equipment parameter early warning alarms, manual task requests, completing flower basket exchanges, and shelf offline; S13: Redis cache: real-time data required for some common production or transportation is recorded in distributed Redis, which can be shared by other systems; S14: Protocol analysis, SCADA system, compatible with TCP / UDP network protocols. At the same time, the application layer protocol is also compatible with Modbus TCP, OPC DA, OPC UA, S7, FINS and many other protocols, and can parse and convert the protocols; S2: Dispatch system is the core system responsible for creating FMS material handling tasks; S21: Dispatch includes task creation and task monitoring and is divided into multiple sub-modules to create algorithms; S3: Create a process, traverse the real-time data of all devices, remove offline devices, select devices that need to be moved in and out, and put them into two groups respectively.

[0007] S31: traverse all the equipment that needs to be moved in, find all the relevant information of the equipment location, feed buffer group, equipment path set, process alignment, exclusive buffer location, MES alignment; S32: Perform a starting point check based on the machines and cache positions found in step S31, put the passed ones into the starting point list, calculate the comprehensive priority of all starting points, and calculate the comprehensive priority algorithm to obtain a comprehensive priority set of the starting points; S33: The calculated path that the starting point and the end point need to pass through is combined with the busy congestion index of the path obtained by the machine learning algorithm to obtain the actual time spent on the path, and the final path priority is obtained by coefficient superposition. Step S31 is executed until all the required points are traversed and a list of all the end points that can select the starting point and the comprehensive priority is obtained; S34: Put all the pairs as input parameters into the Hungarian algorithm function, and return an overall optimal set of starting and ending tasks; S4: Comprehensive priority algorithm, find all the starting machines, the priority of the end machines, the number of outgoing flower baskets, the number of incoming flower baskets, and the outgoing time, and get a preliminary comprehensive priority; S41: Find all cache locations according to the feed temporary storage group and the exclusive cache location, and calculate the preliminary comprehensive priority according to the cache location priority, the flower basket discharge time, and the QTime timeout warning time; S42: Find all the executing tasks that meet the transfer conditions, combine the task status, the priorities of the task start point and the two end points to get the preliminary priority, compare it with the priority of the task itself and remove it, if the transfer priority is high, keep it; S43: Special task check to see if there is any special task requirement for the endpoint. If so, create a special task directly.

[0008] S44: The path congestion, the production rhythm of other machines, and the overdue material discharge time of other machines are obtained from the machine learning results. The priority of the task generated between the two points is calculated by weighted calculation together with the previously calculated priority. S5: The FMS system optimizes task scheduling and execution efficiency by introducing machine learning algorithms and combining real-time data with historical statistical data. The main application points include: equipment portrait establishment, task completion time estimation, continuous learning and optimization; S5: Apply the Hungarian algorithm to efficiently allocate tasks between the main equipment in each process in the FMS to ensure the optimization of resource utilization, including; task and equipment matching modeling, optimal task allocation, dynamic adjustment and real-time optimization; S6: RCS core algorithm and process; S61: The core business process of the RCS (Robot Control System) platform includes four stages: task reception, task allocation, path planning and task execution: S62: Task reception: The RCS platform receives the handling task instructions from the FMS or other upstream systems and parses the task content, including the starting point, end point, and task priority information; S63: Task allocation: RCS reasonably allocates handling tasks according to the real-time status of AGV (position, power, task load) and the estimated task completion time given by the FMS system to ensure optimal resource allocation and avoid task conflicts and no-load operation; S64: Path planning: RCS calculates the optimal driving path of the AGV through the path planning algorithm, taking into account the workshop environment, obstacle distribution, task priority and path congestion to achieve path optimization; S65: Task execution: The AGV executes the task according to the planned path. The RCS monitors the AGV status in real time and dynamically adjusts the path to handle emergencies, such as obstacle avoidance and re-planning the path, to ensure the successful completion of the task. S7: The core algorithms of the RCS platform include task scheduling algorithm, path planning algorithm and dynamic obstacle avoidance algorithm: S8: Deep integration of FMS and RCS; S81: Two-way real-time interaction of task information As the command center of the production process, FMS establishes a two-way real-time data channel with the RCS platform to achieve seamless connection of task information. FMS decomposes production tasks and sends them to RCS, which assists FMS in optimizing task scheduling and resource allocation strategies through task status and execution feedback; S82: Collaborative scheduling based on global optimization FMS integrates the production process, equipment status and logistics requirements of the workshop, and transmits the comprehensive priority calculation results to RCS; RCS generates an efficient task execution plan based on the global optimization strategy of FMS and the status information of AGV, realizing the deep collaboration between workshop logistics and production; S83: Integration of dynamic path and real-time scheduling RCS dynamically adjusts the path planning strategy of AGV according to the production scheduling data and task priority provided by FMS to ensure seamless connection between material transportation and production nodes. FMS timely optimizes the production process and reduces waiting and resource waste through the path congestion information fed back by RCS.

[0009] S84: Data-driven intelligent optimization FMS and RCS share data, combine historical operation data and real-time monitoring information, and use machine learning and algorithm optimization technology to continuously improve scheduling rules and path planning logic. The deeply integrated data analysis capabilities enable the entire system to more accurately predict demand, avoid conflicts and optimize overall production efficiency.

[0010] S85: Collaborative response mechanism for exception handling. When an emergency occurs in the workshop (such as equipment failure, task change or AGV abnormality), FMS and RCS quickly handle the exception through a deeply integrated collaborative response mechanism. FMS adjusts the production plan in a timely manner and sends new task requirements to RCS; RCS dynamically replans the path and scheduling strategy to ensure the continuous and efficient operation of the logistics system. At the same time, through the status feedback mechanism, it assists FMS in optimizing the production process and reducing the impact on the overall production rhythm.

[0011] Preferably, the equipment profile is established for each host equipment, including the equipment's operating status, task completion efficiency, available time period, historical task data, and failure rate parameters. According to the characteristics of different equipment, the task allocation strategy is dynamically adjusted in different time periods (peak / valley) and under different process conditions to improve resource utilization.

[0012] Preferably, the task completion time estimation combines historical statistical data with real-time task data, based on sliding window technology and regression prediction model, and adopts a machine learning model of gradient boosting tree to accurately estimate the completion time of current and future tasks.

[0013] Dynamically adjust model parameters according to different processes, task types and equipment conditions to ensure the accuracy and reliability of estimated time.

[0014] Preferably, the machine continuous learning and optimization system continuously collects task execution time and equipment status data during operation, and updates equipment portraits and prediction models.

[0015] Through the feedback mechanism, the algorithm weights and model accuracy are continuously optimized to improve the efficiency and intelligence level of task scheduling.

[0016] As a preferred method, the task and equipment matching modeling abstracts the task allocation problem between the host equipment of each process into a bipartite graph matching problem; establishes a cost matrix, in which the cost value can represent the execution cost between the task and the equipment, specifically including the following items: AGV docking time, task priority, and task waiting time; The optimal task allocation iteratively searches for the minimum cost matching solution to ensure that the task allocation solution reaches the global optimum. The algorithm can efficiently handle the matching problems of multiple tasks and multiple devices, and the computational complexity is O(n³), which meets the real-time computing requirements. Dynamic adjustment and real-time optimization: During the task execution process, if the device status changes (such as failure, resource occupation), the system can re-run the Hungarian algorithm and dynamically calculate the optimal task allocation plan. Combining the task priority and the real-time status of the device, the system can complete the reallocation in the shortest time, thereby improving the flexibility and stability of task scheduling. Cost Matrix .

[0017] Preferably, the task scheduling algorithm is based on AGV status data (position, power, load), task priority and estimated task completion time, achieves optimal task allocation through a heuristic algorithm, and adopts dynamic task exchange to reduce AGV waiting time and idle rate.

[0018] Preferably, the path planning algorithm adopts A* algorithm and Dijkstra algorithm to calculate the shortest path from the starting point to the end point of the AGV, and combines all global vehicle path data to achieve road right pre-allocation and dynamic re-planning to avoid multi-AGV path conflicts and realize efficient path planning.

[0019] Preferably, the dynamic obstacle avoidance algorithm: based on sensor data and real-time environmental perception, RCS dynamically identifies the location of obstacles, and through local path replanning, ensures that the AGV avoids obstacles and maintains the consistency and safety of task execution.

[0020] As a preferred method, in the task scheduling algorithm, the RCS system introduces a dynamic task scheduling mechanism, which can dynamically adjust the task allocation strategy according to the real-time status of the AGV (including location, power, load) and the task requirements of the workshop. Combined with task priority, path congestion and expected task completion time, it intelligently optimizes resource allocation, reduces the waiting time and empty load rate of AGV, and achieves more efficient logistics scheduling.

[0021] The beneficial effects of the present invention are: the overall handling efficiency of AGV is improved, and the execution time of a single AGV handling task is shortened by 10%; the production capacity of the workshop is improved, and the handling rate exceeds 96% without interference; the logistics cost of the workshop is reduced, and the overall number of AGVs is reduced by 10%; the material handling pass rate is increased by 15%. The deeply integrated data analysis capability enables the entire system to more accurately predict demand, avoid conflicts and optimize overall production efficiency, optimize production processes, reduce waiting and resource waste, and increase the flexibility of path planning and the reliability of task completion. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0023] Figure 1 is a schematic diagram of a SCADA system of a comprehensive algorithm for dispatching and scheduling AGV handling tasks according to an embodiment of the present invention; Figure 2 is a schematic diagram of a Dispatch system of a comprehensive algorithm for dispatching and scheduling AGV handling tasks according to an embodiment of the present invention; Figure 3 is a schematic diagram of a creation process in step S3 according to an embodiment of the present invention; Figure 4 is a schematic diagram of a comprehensive priority algorithm in step S4 according to an embodiment of the present invention; Figure 5 is a schematic diagram of a continuous learning and optimization system according to an embodiment of the present invention; Figure 6 It is a schematic diagram of dynamic adjustment and real-time optimization according to an embodiment of the present invention. DETAILED DESCRIPTION

[0024] To further illustrate each embodiment, the present invention provides drawings, which are part of the disclosure of the present invention and are mainly used to illustrate the embodiments and can be used in conjunction with the relevant descriptions in the specification to explain the operating principles of the embodiments. With reference to these contents, ordinary technicians in the field should be able to understand other possible implementations and advantages of the present invention. The components in the figures are not drawn to scale, and similar component symbols are generally used to represent similar components.

[0025] According to an embodiment of the present invention, a comprehensive algorithm for dispatching and scheduling AGV handling tasks is provided.

[0026] Embodiment 1: like Figure 1-6 As shown, the comprehensive algorithm for dispatching and scheduling AGV handling tasks according to an embodiment of the present invention includes the following steps: S1: SCADA system; equipment communication, data acquisition and storage, protocol analysis, logic processing; S11: The communication protocol, address, and communication rhythm of the equipment are configured through the system. The data is stored in ES (ElasticSearch). The data is kept for one year and backed up regularly. The system implements business logic processing through LUA scripts. S12: Event triggering: set the event triggering conditions through LUA scripts, and push the triggering events to the consumer end through MQ queues for processing, realizing distributed business logic processing, reducing coupling, and being flexible without affecting mainstream business. Such as equipment parameter early warning alarms, manual task requests, completing flower basket exchanges, and shelf offline; S13: Redis cache: real-time data required for some common production or transportation is recorded in distributed Redis, which can be shared by other systems; S14: Protocol analysis, SCADA system, compatible with TCP / UDP network protocols. At the same time, the application layer protocol is also compatible with Modbus TCP, OPC DA, OPC UA, S7, FINS and many other protocols, and can parse and convert the protocols; S2: Dispatch system is the core system responsible for creating FMS material handling tasks; S21: Dispatch includes task creation and task monitoring and is divided into multiple sub-modules to create algorithms; S3: Create a process, traverse the real-time data of all devices, remove offline devices, select devices that need to be moved in and out, and put them into two groups respectively.

[0027] S31: traverse all the equipment that needs to be moved in, find all the relevant information of the equipment location, feed buffer group, equipment path set, process alignment, exclusive buffer location, MES alignment; S32: Perform a starting point check based on the machines and cache positions found in step S31, put the passed ones into the starting point list, calculate the comprehensive priority of all starting points, and calculate the comprehensive priority algorithm to obtain a comprehensive priority set of the starting points; S33: The calculated path that the starting point and the end point need to pass through is combined with the busy congestion index of the path obtained by the machine learning algorithm to obtain the actual time spent on the path, and the final path priority is obtained by coefficient superposition. Step S31 is executed until all the required points are traversed and a list of all the end points that can select the starting point and the comprehensive priority is obtained; S34: Put all the pairs as input parameters into the Hungarian algorithm function, and return an overall optimal set of starting and ending tasks; S4: Comprehensive priority algorithm, find all the starting machines, the priority of the end machines, the number of outgoing flower baskets, the number of incoming flower baskets, and the outgoing time, and get a preliminary comprehensive priority; S41: Find all cache locations according to the feed temporary storage group and the exclusive cache location, and calculate the preliminary comprehensive priority according to the cache location priority, the flower basket discharge time, and the QTime timeout warning time; S42: Find all the executing tasks that meet the transfer conditions, combine the task status, the priorities of the task start point and the two end points to get the preliminary priority, compare it with the priority of the task itself and remove it, if the transfer priority is high, keep it; S43: Special task check to see if there is any special task requirement for the endpoint. If so, create a special task directly.

[0028] S44: The path congestion, the production rhythm of other machines, and the overdue material discharge time of other machines are obtained from the machine learning results. The priority of the task generated between the two points is calculated by weighted calculation together with the previously calculated priority. S5: The FMS system optimizes task scheduling and execution efficiency by introducing machine learning algorithms and combining real-time data with historical statistical data. The main application points include: equipment portrait establishment, task completion time estimation, continuous learning and optimization; S5: Apply the Hungarian algorithm to efficiently allocate tasks between the main equipment in each process in the FMS to ensure the optimization of resource utilization, including; task and equipment matching modeling, optimal task allocation, dynamic adjustment and real-time optimization; S6: RCS core algorithm and process; S61: The core business process of the RCS (Robot Control System) platform includes four stages: task reception, task allocation, path planning and task execution: S62: Task reception: The RCS platform receives the handling task instructions from the FMS or other upstream systems and parses the task content, including the starting point, end point, and task priority information; S63: Task allocation: RCS reasonably allocates handling tasks according to the real-time status of AGV (position, power, task load) and the estimated task completion time given by the FMS system to ensure optimal resource allocation and avoid task conflicts and no-load operation; S64: Path planning: RCS calculates the optimal driving path of the AGV through the path planning algorithm, taking into account the workshop environment, obstacle distribution, task priority and path congestion to achieve path optimization; S65: Task execution: The AGV executes the task according to the planned path. The RCS monitors the AGV status in real time and dynamically adjusts the path to handle emergencies, such as obstacle avoidance and re-planning the path, to ensure the successful completion of the task. S7: The core algorithms of the RCS platform include task scheduling algorithm, path planning algorithm and dynamic obstacle avoidance algorithm: S8: Deep integration of FMS and RCS; S81: Two-way real-time interaction of task information As the command center of the production process, FMS establishes a two-way real-time data channel with the RCS platform to achieve seamless connection of task information. FMS decomposes production tasks and sends them to RCS, which assists FMS in optimizing task scheduling and resource allocation strategies through task status and execution feedback; S82: Collaborative scheduling based on global optimization FMS integrates the production process, equipment status and logistics requirements of the workshop, and transmits the comprehensive priority calculation results to RCS; RCS generates an efficient task execution plan based on the global optimization strategy of FMS and the status information of AGV, realizing the deep collaboration between workshop logistics and production; S83: Integration of dynamic path and real-time scheduling RCS dynamically adjusts the path planning strategy of AGV according to the production scheduling data and task priority provided by FMS to ensure seamless connection between material transportation and production nodes. FMS timely optimizes the production process and reduces waiting and resource waste through the path congestion information fed back by RCS.

[0029] S84: Data-driven intelligent optimization FMS and RCS share data, combine historical operation data and real-time monitoring information, and use machine learning and algorithm optimization technology to continuously improve scheduling rules and path planning logic. The deeply integrated data analysis capabilities enable the entire system to more accurately predict demand, avoid conflicts and optimize overall production efficiency.

[0030] S85: Collaborative response mechanism for exception handling. When an emergency occurs in the workshop (such as equipment failure, task change or AGV abnormality), FMS and RCS quickly handle the exception through a deeply integrated collaborative response mechanism. FMS adjusts the production plan in a timely manner and sends new task requirements to RCS; RCS dynamically replans the path and scheduling strategy to ensure the continuous and efficient operation of the logistics system. At the same time, through the status feedback mechanism, it assists FMS in optimizing the production process and reducing the impact on the overall production rhythm.

[0031] Embodiment 2: like Figure 1-6As shown, the equipment profile is established for each host equipment, including the equipment's operating status, task completion efficiency, available time period, historical task data, and failure rate parameters. According to the characteristics of different equipment, the task allocation strategy is dynamically adjusted in different time periods (peak / valley) and under different process conditions to improve resource utilization.

[0032] The task completion time estimation combines historical statistical data with real-time task data, based on sliding window technology and regression prediction model, and adopts a machine learning model of gradient boosting tree to accurately estimate the completion time of current and future tasks.

[0033] Dynamically adjust model parameters according to different processes, task types and equipment conditions to ensure the accuracy and reliability of estimated time.

[0034] Embodiment three: like Figure 1-6 As shown in the figure, during the operation of the machine continuous learning and optimization system, it continuously collects task execution time and equipment status data, and updates the equipment portrait and prediction model.

[0035] Through the feedback mechanism, the algorithm weights and model accuracy are continuously optimized to improve the efficiency and intelligence level of task scheduling.

[0036] The task and equipment matching modeling abstracts the task allocation problem between the host equipment of each process into a bipartite graph matching problem; establishes a cost matrix, in which the cost value can represent the execution cost between the task and the equipment, specifically including the following items: AGV docking time, task priority, and task waiting time; The optimal task allocation iteratively searches for the minimum cost matching solution to ensure that the task allocation solution reaches the global optimum. The algorithm can efficiently handle the matching problems of multiple tasks and multiple devices, and the computational complexity is O(n³), which meets the real-time computing requirements. Dynamic adjustment and real-time optimization: During the task execution process, if the device status changes (such as failure, resource occupation), the system can re-run the Hungarian algorithm and dynamically calculate the optimal task allocation plan. Combining the task priority and the real-time status of the device, the system can complete the reallocation in the shortest time, thereby improving the flexibility and stability of task scheduling. Cost Matrix .

[0037] Embodiment 4: like Figure 1-6 As shown, the task scheduling algorithm is based on AGV status data (position, power, load), task priority and expected task completion time, achieves optimal task allocation through a heuristic algorithm, and adopts dynamic task exchange to reduce AGV waiting time and idle rate.

[0038] The path planning algorithm adopts the A* algorithm and the Dijkstra algorithm to calculate the shortest path from the starting point to the end point of the AGV, and combines all global vehicle path data to achieve road right pre-allocation, dynamic re-planning, avoid multi-AGV path conflicts, and realize efficient path planning.

[0039] The dynamic obstacle avoidance algorithm: Based on sensor data and real-time environmental perception, RCS dynamically identifies the location of obstacles and ensures that the AGV avoids obstacles and maintains the consistency and safety of task execution through local path replanning.

[0040] In the task scheduling algorithm, the RCS system introduces a dynamic task scheduling mechanism, which can dynamically adjust the task allocation strategy according to the real-time status of the AGV (including location, power, and load) and the task requirements of the workshop. Combined with task priority, path congestion, and expected task completion time, it intelligently optimizes resource allocation, reduces the waiting time and empty load rate of AGV, and achieves more efficient logistics scheduling.

[0041] In summary, with the help of the above technical solutions of the present invention, the overall AGV handling efficiency is improved, and the execution time of the AGV handling task is shortened by 10%; the workshop production capacity is improved, and the handling rate exceeds 96% without interference; the workshop logistics cost is reduced, and the overall number of AGVs is reduced by 10%; the material handling pass rate is increased by 15%. The deeply integrated data analysis capability enables the entire system to more accurately predict demand, avoid conflicts and optimize overall production efficiency, optimize production processes, reduce waiting and resource waste, and increase the flexibility of path planning and the reliability of task completion.

[0042] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A comprehensive algorithm for dispatching and scheduling AGV handling tasks, characterized in that: The steps include: S1: SCADA system; equipment communication, data acquisition and storage, protocol analysis, logic processing; S2: Dispatch system is the core system responsible for creating FMS material handling tasks; S21: Dispatch includes task creation and task monitoring and is divided into multiple sub-modules to create algorithms; S3: Create a process, traverse the real-time data of all devices, remove offline devices, select devices that need to be moved in and out, and put them into two groups respectively; S4: Comprehensive priority algorithm, find all the starting machines, the priority of the end machines, the number of outgoing flower baskets, the number of incoming flower baskets, and the outgoing time, and get a preliminary comprehensive priority; S5: The FMS system optimizes task scheduling and execution efficiency by introducing machine learning algorithms and combining real-time data with historical statistical data. The main application points include: equipment portrait establishment, task completion time estimation, continuous learning and optimization; S5: Apply the Hungarian algorithm to efficiently allocate tasks between the main equipment in each process in the FMS to ensure the optimization of resource utilization, including; task and equipment matching modeling, optimal task allocation, dynamic adjustment and real-time optimization S6: RCS core algorithm and process; S7: The core algorithms of the RCS platform include task scheduling algorithm, path planning algorithm and dynamic obstacle avoidance algorithm: S8: Deep integration of FMS and RCS; S81: Two-way real-time interaction of task information As the command center of the production process, FMS establishes a two-way real-time data channel with the RCS platform to achieve seamless connection of task information. FMS decomposes production tasks and sends them to RCS. RCS assists FMS in optimizing task scheduling and resource allocation strategies through task status and execution feedback. S82: Collaborative scheduling based on global optimization FMS integrates the production process, equipment status and logistics requirements of the workshop, and transmits the comprehensive priority calculation results to RCS; RCS generates an efficient task execution plan based on the global optimization strategy of FMS and the status information of AGV, realizing the deep collaboration between workshop logistics and production; S83: Integration of dynamic path and real-time scheduling RCS dynamically adjusts the path planning strategy of AGV according to the production scheduling data and task priority provided by FMS to ensure the seamless connection between material transportation and production nodes. FMS timely optimizes the production process and reduces waiting and resource waste through the path congestion information fed back by RCS; S84: Data-driven intelligent optimization FMS and RCS share data, combine historical operation data and real-time monitoring information, and use machine learning and algorithm optimization technology to continuously improve scheduling rules and path planning logic. The deeply integrated data analysis capabilities enable the entire system to more accurately predict demand, avoid conflicts and optimize overall production efficiency; S85: Collaborative response mechanism for exception handling. When an emergency occurs in the workshop, FMS and RCS quickly handle the exception through a deeply integrated collaborative response mechanism. FMS adjusts the production plan in a timely manner and issues new task requirements to RCS. RCS dynamically replans routes and scheduling strategies to ensure the continuous and efficient operation of the logistics system. At the same time, through the status feedback mechanism, it assists FMS in optimizing the production process and reducing the impact on the overall production rhythm.

2. The comprehensive algorithm for dispatching and scheduling AGV handling tasks according to claim 1, characterized in that: S11: The communication protocol, address, and communication rhythm of the equipment are configured by the system. The data is stored in ES. The data is saved for one year and backed up regularly. The system implements business logic processing through LUA scripts. S12: Event triggering; set the event triggering conditions through LUA scripts, and push the triggering events to the consumer end through the MQ queue for processing, realizing distributed business logic processing, reducing coupling, and being flexible without affecting mainstream business, such as equipment parameter early warning alarms, manual request tasks, completing flower basket exchange, and shelf offline S13: Redis cache: real-time data required for some common production or transportation is recorded in distributed Redis, which can be shared by other systems; S14: Protocol analysis, SCADA system, compatible with TCP / UDP network protocol, and the application layer protocol is also compatible with Modbus TCP, OPC DA, OPC UA, S7, FINS and many other protocols, which can parse and convert the protocol; The equipment portrait is established for each host equipment, including the equipment's operating status, task completion efficiency, available time period, historical task data, and failure rate parameters. According to the characteristics of different equipment, the task allocation strategy is dynamically adjusted in different time periods and under different process conditions to improve resource utilization.

3. The comprehensive algorithm for dispatching and scheduling AGV handling tasks according to claim 2 is characterized in that S3 1: Traverse all the equipment that needs to be moved in, find all the relevant information of the equipment's location, feed buffer group, equipment path set, process alignment, exclusive buffer location, and MES alignment; S32: Perform a starting point check based on the machines and cache positions found in step S31, put the passed ones into the starting point list, calculate the comprehensive priority of all starting points, and calculate the comprehensive priority algorithm to obtain a comprehensive priority set of the starting points; S33: The calculated path that the starting point and the end point need to pass through is combined with the busy congestion index of the path obtained by the machine learning algorithm to obtain the actual time spent on the path, and the final path priority is obtained by coefficient superposition. Step S31 is executed until all the required points are traversed and a list of all the end points that can select the starting point and the comprehensive priority is obtained; S34: Put all the pairs as input parameters into the Hungarian algorithm function, and return an overall optimal set of starting and ending tasks; The task completion time estimation combines historical statistical data with real-time task data, based on sliding window technology and regression prediction model, and adopts the machine learning model of gradient boosting tree to accurately estimate the completion time of current and future tasks; Dynamically adjust model parameters for different processes, task types and equipment conditions to ensure the accuracy and reliability of estimated time.

4. The comprehensive algorithm for dispatching and scheduling AGV handling tasks according to claim 3 is characterized in that: S41: Find all cache locations according to the feed temporary storage group and the exclusive cache location, and calculate the preliminary comprehensive priority according to the cache location priority, the flower basket discharge time, and the QTime timeout warning time; S42: Find all the executing tasks that meet the transfer conditions, combine the task status, the priorities of the task start point and the two end points to get the preliminary priority, compare it with the priority of the task itself and remove it, if the transfer priority is high, keep it; S43: Check special tasks to see if there are any special task requirements for the endpoint. If so, directly create a special task. S44: The path congestion, the production rhythm of other machines, and the overdue material discharge time of other machines are obtained from the machine learning results. The priority of the task generated between the two points is calculated by weighted calculation together with the previously calculated priority. During operation, the machine continuous learning and optimization system continuously collects task execution time and equipment status data, and updates equipment portraits and prediction models; Through the feedback mechanism, the algorithm weights and model accuracy are continuously optimized to improve the efficiency and intelligence level of task scheduling.

5. The comprehensive algorithm for dispatching and scheduling AGV handling tasks according to claim 4 is characterized in that: S61: The core business process of the RCS platform includes four stages: task reception, task allocation, path planning and task execution: S62: Task reception: The RCS platform receives the handling task instructions from the FMS or other upstream systems and parses the task content, including the starting point, end point, and task priority information; S63: Task allocation: RCS reasonably allocates handling tasks according to the real-time status of AGV and the estimated task completion time given by the FMS system to ensure optimal resource allocation and avoid task conflicts and no-load operation; S64: Path planning: RCS calculates the optimal driving path of the AGV through the path planning algorithm, taking into account the workshop environment, obstacle distribution, task priority and path congestion to achieve path optimization; S65: Task execution: AGV executes the task according to the planned path. RCS monitors the AGV status in real time and dynamically adjusts the path to handle emergencies, such as obstacle avoidance and re-planning the path, to ensure the successful completion of the task. The task and equipment matching modeling abstracts the task allocation problem between the host equipment of each process into a bipartite graph matching problem; establishes a cost matrix, in which the cost value can represent the execution cost between the task and the equipment, specifically including the following items: AGV docking time, task priority, and task waiting time; The optimal task allocation iteratively searches for the minimum cost matching solution to ensure that the task allocation solution reaches the global optimum. The algorithm can efficiently handle the matching problems of multiple tasks and multiple devices, and the computational complexity is O (n³), which meets the real-time computing requirements. Dynamic adjustment and real-time optimization: During the task execution process, if the device status changes, the system can re-run the Hungarian algorithm to dynamically calculate the optimal task allocation plan. Combining the task priority and the real-time status of the device, the system can complete the reallocation in the shortest time, thereby improving the flexibility and stability of task scheduling. Cost Matrix .

6. The comprehensive algorithm for dispatching and scheduling AGV handling tasks according to claim 5 is characterized in that: The task scheduling algorithm: based on AGV status data, task priority and task estimated completion time, achieves optimal task allocation through a heuristic algorithm, adopts dynamic task exchange, and reduces AGV waiting time and idle rate.

7. The comprehensive algorithm for dispatching and scheduling AGV handling tasks according to claim 6, characterized in that: The path planning algorithm adopts the A* algorithm and the Dijkstra algorithm to calculate the shortest path from the starting point to the end point of the AGV, and combines all global vehicle path data to achieve road right pre-allocation, dynamic re-planning, avoid multi-AGV path conflicts, and realize efficient path planning.

8. The comprehensive algorithm for dispatching and scheduling AGV handling tasks according to claim 7, characterized in that: The dynamic obstacle avoidance algorithm: Based on sensor data and real-time environmental perception, RCS dynamically identifies the location of obstacles and ensures that the AGV avoids obstacles and maintains the consistency and safety of task execution through local path replanning.

9. The comprehensive algorithm for dispatching and scheduling AGV handling tasks according to claim 8, characterized in that: In the task scheduling algorithm, the RCS system introduces a dynamic task scheduling mechanism, which can dynamically adjust the task allocation strategy according to the real-time status of the AGV and the task requirements of the workshop, and intelligently optimize resource allocation based on task priority, path congestion, and expected task completion time, thereby reducing the waiting time and idle rate of the AGV and achieving more efficient logistics scheduling.

Images