Research Of Automated Guided Vehicle System Path Optimization Based On Time Window

Automated guided vehicle(AGV) is wheeled mobile robot which can realize material transportation in flexible manufacture system(FMS) and stereoscopic warehouse(SW). Due to the characteristic of highly flexibility and easy to manage, multiple automated guided vehicles system as the key subsystem of logistics system(LS) and flexible manufacture system is increasingly applied in the delivery material. Path optimization is one of the key technologies in the application of multi AGV system, this technology is hot spot and difficult to study, so the research has important theoretical and practical significance.In this thesis, multiple AGV system path optimization and the strategy of coordination and conflict-free between AGV are investigated based on domestic and foreign research works, the main research work is as follow:Firstly, taking the AGV with fixed routing guidance as the research object, establishing the working environment of AGV for material transportation, and the map model is built with topology structure under the condition of the known information of AGV running environment. The fixed time constant τ is added to the each node in AGV moving path as punishment when the optimal path of single AGV is searched, the more turning, the more cost. And taking the shortest time as evaluation index, the shortest paths of the former K of single AGV are searched off-line by the Dijsktra algorithm.Secondly, the the shortest paths of the former K are the initial paths, the way of combination and superposition is proposed. Then the shortest paths of the former K are combined and compared respectively with the paths of AGV which is running through the process of the arrangement of ideal time window, update time window with the way of combination and superposition and conflict detection. The conflicts are divided into different types and different resolution strategies are adoped to solve each conflict, and taking the minimum sum of AGV running time as evaluation index, then the optimization of the collision free path based on time window for multi AGV is implemented in the off-line state. At the same time, the detection of the sudden condition of AGV is completed by on-line detection method after the completion of path optimization offline based on time window. The coordination online for multiple AGV system is achieved by the strategy of priority changing of AGV passing the node and waiting strategy.Finally, on the basis of theoretical research, the simulation platform of multiple AGV ground control system is established by Visual Studio2010 and Visual C++ programming language. The path search of single AGV and the conflict-free path optimization of multiple AGV system are tested respectively. Through the simulation results, the validity and reliability of the proposed algorithm in the thesis is verified. The algorithm and strategy can improve system flexibility and efficiency, and can be the theory basis and work instruction for the application of multiple AGV system in realistic environment.