The Research On Path Planning For Multi-automated Guided Vehicles System

With the improvement of the level of production logistics automation productionlogistic and the development of flexible manufacturing systems and automaticthree-dimensional warehouses, as the critical subsystems of the logistics systems andflexible manufacturing systems, the applications automatic guided vehicle （AGV）systems are more and used more widely. Automatic guided vehicle systems aremainly used for material transport links, it can facilitate automated management andimprove the flexibility of manufacturing systems and production efficiency. The pathplanning and optimization of the AGV system is the key research problem. The AGVpath planning system mainly includes Single AGV optimal path planning andmulti-AGVs’ coordination and avoid collision. Along with the rising number of AGVin the system, the AGVs path planning is becoming more and more complex. We areresearching the path search and optimization based on the research by domestic andabroad scholars.This paper describes the structure and system composition of the AGV, and thenanalysis the main problems of its control system, describes the navigation modes’direct method, finally decide the research purpose-laser guidance. in addition, we doresearch on Single AGV optimal path planning, optimize the Dijkstra algorithm bycreating an electronic map. Improve search efficiency by narrow the searching area.Add evaluation indicator to make sure the optimized searched path is more suitable tothe actual operation, so that the single AGV’s path planning is achieve. Use the timewindow based on pre-planning algorithm, and put forward the concept of stretch timewindow, for a continuous path and continuous time, the path can be searched directly;for the continuous path and discontinuous time, the stretch time window can providemore time period to search for ways. Due to the conflicts that the introduction of flextime window makes in some nodes of the path planning, this paper adopts twostrategies to solve:1) search the path again, avoid the note that conflict with the timewindow;2) Adjust the translation time window quickly to get collision free pathplanning of AGV. Combining the improved Dijkstra algorithm and time window, planthe AGV path in priority order. Detecting whether the follow-up planning path hasSpace and time conflict with the existing planning path, and call optimizationalgorithm and avoid Strategy for the choice of the optimal path, so as to realize thecollision free path planning of AGV; Finally developed the AGV system platform, theground control simulation system for single and multiple sets of AGV path planningoptimization algorithm simulation verification. The results show that the optimizedpath planning algorithm and conflict resolution strategy is feasible, this lays a certainfoundation for practical applications.