Research On Path Planning And Scheduling Mechanism Of Multi-AGV System In Automation Factory

With the rapid development of industrial automation technology,automated factories have used a large number of robots instead of labor in some complex,cumbersome and physically demanding tasks.The automatic transfer vehicle(AGV)is an important branch of the automated factory system transportation.Its design and research and development meet the requirements of the development of intelligent industrial systems,which greatly promotes the technological upgrade of the factory,reduces the labor demand of the enterprise,and saves the labor cost of the factory.Based on the background of automated factories,this paper studies the path planning and scheduling mechanism in the process of AGV carrying out transportation tasks,so as to further reduce transportation costs and improve system operation efficiency.The main research contents are as follows:Firstly,aiming at the task allocation problem of multiple AGV systems,this paper adopts the multi-task execution mode according to the characteristics of task allocation,and establishes a task allocation model with cargo weight,AGV power and AGV maximum load as constraints.In view of the low efficiency of solving the problem at present,the genetic algorithm is optimized.Through simulation and comparison,the optimized genetic algorithm has a better optimization effect than the particle swarm algorithm.Secondly,aiming at the path planning problem,according to the transportation characteristics of the automated factory,the topology grid method is used to model the environment of the actual operating scene.Aiming at the problems of slow solution speed and many path turns in current planning algorithms,an improved A* algorithm is proposed.This algorithm uses a two-way search strategy and a method of weighting AGV turns,which speeds up the calculation and reduces the number of path turns.The simulation shows that the path planning method proposed in this paper can search fewer nodes in the search process and search shorter paths.Thirdly aiming at the problem of path resource conflicts during the operation of multiple AGV systems,the problem is divided into problems under normal driving conditions and problems under abnormal driving conditions.Aiming at path conflicts under normal driving conditions,such as facing conflicts,co-directional conflicts,and node occupancy conflicts,the time window method is used to resolve conflicts.For problems under abnormal driving conditions,such as the grid point occupation problem caused by AGV failure,and the path deadlock caused by the sudden appearance of obstacles in the path,the method of path replanning is used to resolve.The reliability of the algorithm is verified by simulation.Finally,the software of multi-AGVs control system is developed based on Visual Studio software.Taking the raspberry AGV as the experimental platform,the basic functions of single and multiple AGVs and the functions of the scheduling software are tested to verify the correctness and feasibility of the above algorithms and mechanisms.