Multi-AGV decentralized scheduling system based on distributed PLC group

Abstract

The invention discloses a multi-AGV decentralized scheduling system based on a distributed PLC group, and belongs to the technical field of automatic logistics scheduling. The system comprises a distributed PLC control group, a plurality of AGVs, a communication network and an upper task management system. The distributed PLC control group is composed of a plurality of PLC controllers which are in peer-to-peer interconnection through the industrial Ethernet, and the controllers run the same collaborative scheduling logic program and comprise task negotiation, resource map management, conflict detection and resolution and communication interface modules. After an upper system issues a task, through task broadcast auction, distributed path reservation and a conflict cooperation resolution process, autonomous cooperation operation of multiple AGVs under the condition of no central controller is realized. According to the method, the problems of single-point bottleneck and vulnerability of a centralized scheduling system are effectively solved, and the overall scheduling efficiency, the operation stability and the expandability of the multi-AGV system in a dynamic complex environment are remarkably improved through a distributed decision-making and cooperation mechanism.

Description

A decentralized scheduling system for multiple AGVs based on distributed PLC clusters Technical Field

[0001] This invention relates to the field of automation control and logistics scheduling technology, and more specifically, to a decentralized scheduling system for multiple AGVs based on a distributed PLC group. Background Technology

[0002] With the rapid development of intelligent manufacturing and smart logistics, Automated Guided Vehicles (AGVs) have become key equipment for material handling in warehousing and production lines due to their automation and flexibility. Traditional AGV scheduling often adopts a centralized architecture, relying on a single central controller for global task allocation and path planning. As the scale of operations expands and the complexity of tasks increases, this architecture places extremely high demands on the computing power of the central node and is prone to system crashes due to single points of failure. It is difficult to meet the high concurrency and high real-time scheduling requirements of large-scale complex scenarios, and it is prone to path conflicts and traffic congestion.

[0003] To improve system robustness and scalability, distributed scheduling schemes have become a research hotspot. While some existing distributed methods reduce central dependency, they still have shortcomings in areas such as global task optimization and allocation, multi-agent conflict resolution, and real-time adaptation to dynamic environments. For example, some schemes lack effective global resource coordination mechanisms, easily leading to uneven task allocation and resource waste; path planning algorithms struggle to efficiently integrate real-time dynamic information in a distributed environment, causing AGVs to detour or wait, limiting overall operational efficiency; and the system's collaborative avoidance and recovery capabilities are weak when dealing with node failures or sudden obstacles. Therefore, a multi-AGV distributed scheduling system that combines efficient collaboration, dynamic adaptation, and high reliability is urgently needed. Summary of the Invention

[0004] To address the aforementioned technical problems in related technologies, this invention proposes a decentralized scheduling system for multiple AGVs based on a distributed PLC group, which can overcome the above-mentioned shortcomings of the prior art.

[0005] To achieve the above technical objectives, the technical solution of the present invention is implemented as follows: a decentralized scheduling system for multiple AGVs based on a distributed PLC group; the decentralized scheduling system for multiple AGVs based on a distributed PLC group includes: a distributed PLC control group, multiple AGVs, a communication network, and a host task management system; the distributed PLC control group consists of multiple PLC controllers deployed in different physical areas of the working environment, and each PLC controller forms a peer-to-peer communication network through an industrial Ethernet, without a central master-slave relationship; each PLC controller is configured with the same collaborative scheduling logic program, which includes a task negotiation module, a resource map management module, a conflict detection and resolution module, and an AGV communication interface module; the AGV connects to the PLC controller in its current area via wireless communication to receive movement and operation instructions and report its own status, location, and task progress; the host task management system is used to issue transport task orders to any one or more PLC controllers in the distributed PLC control group; the system is configured to execute a scheduling process including task issuance and reception, task broadcasting and bidding, task negotiation and assignment, distributed path planning and reservation, conflict collaborative resolution, and dynamic load balancing.

[0006] Furthermore, the collaborative scheduling logic program is programmed based on the IEC 61131-3 standard and adopts a distributed consensus mechanism to ensure consensus is reached when a node fails. It uses an "event-driven + state machine" model to synchronize data flow and state.

[0007] Furthermore, the AGV is equipped with a wireless communication module, a navigation and positioning module, and an on-board controller, wherein the on-board controller is an embedded PLC or an industrial microcontroller.

[0008] Furthermore, during the task broadcasting and bidding process, PLC controllers communicate peer-to-peer using an industrial communication protocol based on a publish / subscribe model, and execute a multi-objective optimization bidding algorithm to calculate the bid value. The multi-objective optimization bidding algorithm comprehensively considers factors such as the remaining power of the AGV, the urgency of the task, the degree of path congestion, and historical execution efficiency. The task negotiation and assignment process selects the PLC controller to execute the task based on the optimal principle.

[0009] Furthermore, the optimal principle is to minimize the total travel distance of the AGV performing the task to complete the transport task.

[0010] Furthermore, during the distributed path planning and reservation process, the AGV is instructed to perform initial path planning based on the global resource map and initiate resource reservation requests to the PLC controllers to which the path nodes belong along the way. The global resource map is a logical map, which is jointly maintained and updated in real time by all PLC controllers. Each PLC controller holds all map information and has the right to arbitrate and allocate local path resources under its jurisdiction.

[0011] Furthermore, during the conflict resolution process, when multiple AGVs request conflicting resources from the same PLC controller, the conflict detection and resolution module of the PLC controller performs local arbitration based on preset arbitration rules and notifies the relevant AGVs and their management PLC controllers to adjust their paths or speeds through the communication network; the arbitration rules include priority, task urgency, and waiting time.

[0012] Furthermore, during the dynamic load balancing process, each PLC controller continuously monitors the AGV density and task queue in its area. When the load exceeds the threshold, it guides new tasks or AGVs in its area to adjacent areas with lighter loads through the communication network.

[0013] Furthermore, it also includes an incremental map update module for broadcasting update information when environmental obstacles are detected; and a path replanning module for enabling the AGV to make real-time path adjustments based on the updated map, combined with spatiotemporal constraints and rolling optimization strategies.

[0014] Furthermore, the system also includes a fault-tolerant mechanism, in which the AGVs under the jurisdiction of a PLC controller automatically migrate to a nearby PLC controller when a PLC controller fails; and / or, the communication network is an industrial Ethernet with deterministic and real-time capabilities.

[0015] The beneficial effects of this invention are as follows: By constructing a distributed PLC control group based on peer-to-peer network communication and integrating collaborative scheduling logic, multiple AGVs can achieve decentralized autonomous task negotiation, global resource collaborative management, and intelligent resolution of local conflicts, thereby significantly improving the overall scheduling stability, operating efficiency, and dynamic environmental adaptability of the system. At the same time, it effectively enhances the scalability and fault tolerance of the system in large-scale complex scenarios. Detailed Implementation

[0016] The technical solutions of the present invention will be clearly and completely described below with reference to the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention are within the scope of protection of the present invention.

[0017] According to an embodiment of the present invention, a decentralized scheduling system for multiple AGVs based on a distributed PLC group includes: a distributed PLC control group, multiple AGVs, a communication network, and a host task management system. The distributed PLC control group consists of multiple PLC controllers deployed in different physical areas of the working environment. Each PLC controller forms a peer-to-peer communication network via an industrial Ethernet network, without a central master-slave relationship. Each PLC controller is configured with the same collaborative scheduling logic program, which includes a task negotiation module, a resource map management module, a conflict detection and resolution module, and an AGV communication interface module. The AGVs are connected to the PLC controllers in their current area via wireless communication to receive movement and operation instructions and report their own status, location, and task progress. The host task management system is used to issue transport task orders to any one or more PLC controllers in the distributed PLC control group. The system is configured to execute a scheduling process including task issuance and reception, task broadcasting and bidding, task negotiation and assignment, distributed path planning and reservation, conflict collaborative resolution, and dynamic load balancing.

[0018] According to an embodiment of the present invention, a decentralized scheduling system for multiple AGVs based on a distributed PLC group is provided. In a specific implementation, the collaborative scheduling logic program is programmed based on the IEC 61131-3 standard and a distributed consensus mechanism is adopted to ensure consensus is reached when a node fails. An "event-driven + state machine" model is used to synchronize data flow and state.

[0019] According to an embodiment of the present invention, a decentralized scheduling system for multiple AGVs based on a distributed PLC group is provided. In a specific embodiment, the AGV is equipped with a wireless communication module, a navigation and positioning module, and an on-board controller. The on-board controller is an embedded PLC or an industrial microcontroller.

[0020] According to an embodiment of the present invention, a decentralized scheduling system for multiple AGVs based on a distributed PLC group, in a specific implementation, during the task broadcasting and bidding process, the PLC controllers communicate peer-to-peer using an industrial communication protocol based on a publish / subscribe model, and execute a multi-objective optimization bidding algorithm to calculate the bid value. The multi-objective optimization bidding algorithm comprehensively considers factors such as the remaining power of the AGV, the urgency of the task, the degree of path congestion, and historical execution efficiency. The task negotiation and assignment process selects the PLC controller to execute the task based on the optimal principle.

[0021] According to an embodiment of the present invention, a decentralized scheduling system for multiple AGVs based on a distributed PLC group, in a specific implementation, the optimal principle is to minimize the total travel distance of the AGVs performing the task to complete the transport task.

[0022] According to an embodiment of the present invention, a decentralized scheduling system for multiple AGVs based on a distributed PLC group, in a specific implementation, during the distributed path planning and reservation process, the assigned AGV performs initial path planning based on a global resource map and initiates resource reservation requests to the PLC controllers to which the path nodes belong along the way; the global resource map is a logical map, which is jointly maintained and updated in real time by all PLC controllers, and each PLC controller holds all map information and has the right to arbitrate and allocate local path resources under its jurisdiction.

[0023] According to an embodiment of the present invention, a decentralized scheduling system for multiple AGVs based on a distributed PLC group, in a specific implementation, during the conflict resolution process, when multiple AGVs request conflicting resources from the same PLC controller, the conflict detection and resolution module of the PLC controller performs local arbitration based on preset arbitration rules, and notifies the relevant AGVs and their management PLC controllers to adjust their paths or speeds through the communication network; the arbitration rules include priority, task urgency, and waiting time.

[0024] According to an embodiment of the present invention, a decentralized scheduling system for multiple AGVs based on a distributed PLC group is provided. In a specific implementation, during the dynamic load balancing process, each PLC controller continuously monitors the AGV density and task queue in its area. When the load exceeds a threshold, the system guides new tasks or AGVs in the area to adjacent areas with lighter loads through the communication network.

[0025] According to an embodiment of the present invention, a decentralized scheduling system for multiple AGVs based on a distributed PLC group, in a specific implementation, further includes an incremental map update module for broadcasting update information when environmental obstacles are detected; and a path replanning module for enabling the AGVs to perform real-time path adjustments based on the updated map, combined with spatiotemporal constraints and rolling optimization strategies.

[0026] According to an embodiment of the present invention, a decentralized scheduling system for multiple AGVs based on a distributed PLC group is provided. In a specific embodiment, the system further includes a fault tolerance mechanism, wherein when a PLC controller fails, the AGVs under its jurisdiction automatically migrate to a nearby PLC controller; and / or, the communication network is an industrial Ethernet network with deterministic and real-time capabilities.

[0027] To facilitate understanding of the above technical solutions of the present invention, the following detailed description of the above technical solutions of the present invention will be provided through specific usage methods.

[0028] In practical use, the multi-AGV decentralized scheduling system based on a distributed PLC group according to the present invention includes a distributed PLC control group, multiple AGVs, a communication network, and a host task management system.

[0029] The distributed PLC control group consists of multiple PLC controllers deployed in different physical areas of the operating environment. These controllers form a peer-to-peer communication network via industrial Ethernet, without a central master-slave relationship. Each PLC controller is configured with the same collaborative scheduling logic program. This logic program is programmed based on the IEC 61131-3 standard and can be further verified using a Petri network model to ensure logical rigor and deadlock-free characteristics. The collaborative scheduling logic program includes a task negotiation module, a resource map management module, a conflict detection and resolution module, and an AGV communication interface module. Deep collaborative scheduling among the modules is achieved. This logic program employs a distributed consensus mechanism to ensure consistency even when node failures occur in the PLC network. The data flow and state synchronization mechanism between modules adopts an "event-driven + state machine" model to ensure the real-time performance and consistency of scheduling decisions.

[0030] The AGV connects to the PLC controller of its current location via wireless communication, receives movement and operation commands, and reports its own status, location, and task progress. The AGV is equipped with at least a wireless communication module, a navigation and positioning module, and an onboard controller. The onboard controller can be an embedded PLC or an industrial microcontroller, used to execute commands from the managing PLC and control the movement of the AGV itself.

[0031] The host task management system is used to issue handling task orders to any one or more PLCs in the distributed PLC group. The system's scheduling process includes steps such as task issuance and reception, task broadcasting and bidding, task negotiation and assignment, distributed path planning and reservation, conflict resolution, and dynamic load balancing.

[0032] In the task broadcasting and bidding process, PLC controllers communicate peer-to-peer using industrial communication protocols based on a publish / subscribe model, such as OPC UA PubSub or a specific MQTT-based industrial protocol. The system designs and applies a multi-objective optimization bidding algorithm, which comprehensively considers factors such as the AGV's remaining power, task urgency, path congestion, and historical execution efficiency, calculating bid values ​​in a formulaic manner. In the task negotiation and assignment process, PLC-A, receiving the task, collects all bids, selects PLC-B for execution based on the optimal principle, and broadcasts the assignment result for confirmation.

[0033] In the distributed path planning and reservation step, the assigned AGV performs initial path planning based on the global resource map and initiates resource (path segment) reservation requests to the PLCs belonging to the path nodes along the way. The global resource map is a logical map, jointly maintained by all PLCs. Each PLC holds all map information but only has the right to arbitrate and allocate local path resources under its jurisdiction. Map updates are synchronized in real time through the PLC network.

[0034] In the conflict resolution process, if multiple AGVs request conflicting resources from the same PLC, the PLC's conflict detection and resolution module performs local arbitration based on preset rules and notifies the relevant AGVs and their management PLCs to adjust their paths or speeds via the PLC network. This process occurs in parallel across multiple PLCs.

[0035] In the dynamic load balancing process, each PLC continuously monitors the AGV density and task queue in its area. When the load exceeds the threshold, new tasks or AGVs in the area can be guided to adjacent areas with lighter loads through the PLC network.

[0036] The system also includes an incremental map update module that broadcasts obstacle coordinates when sensors detect sudden environmental changes. Upon receiving the updated data, each AGV can use methods such as combining the RRT* algorithm with a spatiotemporal grid map and introducing a "time window reservation mechanism" for path planning to avoid conflicts. Through rolling optimization and replanning strategies, real-time smooth path adjustments are achieved in dynamic environments; for example, the search step size can be automatically reduced to 5cm in narrow areas to ensure path feasibility. The system can also introduce regional grouping and federated learning mechanisms to support distributed learning and decision-making in large-scale AGV clusters. To enhance reliability, the system can add a "heartbeat detection + master / slave failover" mechanism. When a PLC fails, the AGVs under its jurisdiction can automatically migrate to a nearby PLC, ensuring continuous system operation.

[0037] In one specific embodiment, the distributed PLC control group consists of five PLC controllers deployed in different physical areas of the operating environment. The collaborative scheduling logic program configured for each PLC controller is programmed based on the IEC 61131-3 standard and can be designed and verified using a Petri net model. The system includes five AGVs, numbered AGV1 to AGV5. Each AGV's onboard controller is an embedded PLC.

[0038] In this embodiment, the task broadcasting and bidding processes utilize the OPC UA PubSub industrial communication protocol for peer-to-peer communication. Each PLC calculates its bid value based on a greedy algorithm, taking into account the status, location, and area load of the AGVs it manages, and then feeds it back. Task negotiation and assignment select the executing PLC based on the principle of minimizing the total walking distance for the robot to complete the task. Path planning, conflict resolution, and load balancing are consistent with the aforementioned technical solutions, and incremental map updates, rolling optimization, and fault switching mechanisms can also be applied.

[0039] In summary, by utilizing the above-mentioned technical solution of this invention, a distributed PLC control group based on peer-to-peer network communication is constructed and collaborative scheduling logic is integrated, enabling multiple AGVs to achieve decentralized autonomous task negotiation, global resource collaborative management, and intelligent resolution of local conflicts. This significantly improves the overall scheduling stability, operating efficiency, and dynamic environmental adaptability of the system, while effectively enhancing the system's scalability and fault tolerance in large-scale complex scenarios.

[0040] 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 principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A decentralized scheduling system for multiple AGVs based on a distributed PLC cluster, characterized in that, include: The system comprises a distributed PLC control group, multiple AGVs, a communication network, and a host task management system. The distributed PLC control group consists of multiple PLC controllers deployed in different physical areas of the work environment. Each PLC controller forms a peer-to-peer communication network via an industrial Ethernet network, without a central master-slave relationship. Each PLC controller is configured with the same collaborative scheduling logic program, which includes a task negotiation module, a resource map management module, a conflict detection and resolution module, and an AGV communication interface module. The AGVs connect wirelessly to the PLC controllers in their current area to receive movement and operation instructions and report their status, location, and task progress. The host task management system issues transport task orders to any one or more PLC controllers in the distributed PLC control group. The system is configured to execute a scheduling process including task issuance and reception, task broadcasting and bidding, task negotiation and assignment, distributed path planning and reservation, conflict resolution, and dynamic load balancing.

2. The decentralized scheduling system for multiple AGVs based on a distributed PLC group according to claim 1, characterized in that, The collaborative scheduling logic program is programmed based on the IEC 61131-3 standard and uses a distributed consensus mechanism to ensure consensus in the event of node failure. It adopts an "event-driven + state machine" model to achieve data flow and state synchronization.

3. The decentralized scheduling system for multiple AGVs based on a distributed PLC group according to claim 1, characterized in that, The AGV is equipped with a wireless communication module, a navigation and positioning module, and an on-board controller, which is an embedded PLC or an industrial microcontroller.

4. The decentralized scheduling system for multiple AGVs based on a distributed PLC group according to claim 1, characterized in that, During the task broadcasting and bidding process, PLC controllers communicate peer-to-peer using an industrial communication protocol based on a publish / subscribe model, and execute a multi-objective optimization bidding algorithm to calculate the bid value. The multi-objective optimization bidding algorithm comprehensively considers factors such as the remaining power of the AGV, the urgency of the task, the degree of path congestion, and historical execution efficiency. The task negotiation and assignment process selects the PLC controller to execute the task based on the optimal principle.

5. A decentralized scheduling system for multiple AGVs based on a distributed PLC group according to claim 4, characterized in that, The optimal principle is to minimize the total travel distance of the AGV performing the task to complete the transport task.

6. A decentralized scheduling system for multiple AGVs based on a distributed PLC group according to claim 1, characterized in that, During the distributed path planning and reservation process, the AGV is instructed to perform initial path planning based on the global resource map and initiate resource reservation requests to the PLC controllers of the path nodes along the way. The global resource map is a logical map, which is jointly maintained and updated in real time by all PLC controllers. Each PLC controller holds all map information and has the right to arbitrate and allocate local path resources under its jurisdiction.

7. A decentralized scheduling system for multiple AGVs based on a distributed PLC group according to claim 1, characterized in that, During the conflict resolution process, when multiple AGVs request conflicting resources from the same PLC controller, the conflict detection and resolution module of the PLC controller performs local arbitration based on preset arbitration rules and notifies the relevant AGVs and their management PLC controllers to adjust their paths or speeds through the communication network; the arbitration rules include priority, task urgency, and waiting time.

8. A decentralized scheduling system for multiple AGVs based on a distributed PLC group according to claim 1, characterized in that, During the dynamic load balancing process, each PLC controller continuously monitors the AGV density and task queue in its area. When the load exceeds the threshold, it guides new tasks or AGVs in the area to adjacent areas with lighter loads through the communication network.

9. A decentralized scheduling system for multiple AGVs based on a distributed PLC group according to claim 1, characterized in that, It also includes an incremental map update module, which broadcasts update information when environmental obstacles are detected; And a path replanning module, which enables the AGV to make real-time path adjustments based on the updated map, combined with spatiotemporal constraints and rolling optimization strategies.

10. A decentralized scheduling system for multiple AGVs based on a distributed PLC group according to claim 1, characterized in that, The system also includes a fault-tolerant mechanism, in which the AGVs under the jurisdiction of a PLC controller automatically migrate to a nearby PLC controller when a PLC controller fails; and / or, the communication network is an industrial Ethernet with deterministic and real-time capabilities.