Research And Application Of AGV Path Planning And Autonomous Obstacle Avoidance

With the rapid development of science and technology,AGV(automatic guided vehicle),which has the advantages of automation,safety,reliability and flexibility,plays an increasingly important role in intelligent inspection,warehousing and logistics.The key problem in the application of AGV system is path planning and dynamic obstacle avoidance,so it is of great practical significance to study the technology of path planning and dynamic obstacle avoidance algorithm.In this paper,the traditional AGV path planning and the traditional algorithm of autonomous obstacle avoidance are deeply studied,the ant colony algorithm is optimized and the dynamic window method is improved by introducing fuzzy neural network,so as to improve the operation efficiency and reliability of the robot.At the same time,the actual effect of the algorithm is verified by the robot walking and automatic obstacle avoidance experiment controlled by ROS system.First of all,the ROS(robot operating system)system is studied.The system has the characteristics of open source distributed operating system,which can solve the problem of code duplication and poor compatibility when developing robots,and replace the follow-up research and development.Therefore,in this paper,ROS is selected as the software system of mobile robot to carry out the experiment of mobile robot.Secondly,to solve the problems of slow convergence and easy to fall into local optimum in ACO algorithm,the improved Artificial Potential Field algorithm is combined with ACO to reduce the blindness of initial planning.Using the evaluation function and path turning angle of A* algorithm,the paper introduces heuristic information increasing function,improves pheromone updating mechanism and path evaluation function to improve ant colony algorithm.The simulation results show that the improved algorithm promotes the convergence speed and the optimal solution.Then,based on the research of autonomous obstacle avoidance algorithm of mobile robot,the traditional dynamic window algorithm is optimized.The track evaluation function of dynamic window method is improved to give full play to the optimality of global path.At the same time,combining with fuzzy neural network,the weight of evaluation function of dynamic window method is adjusted dynamically according to the environmental information to improve the planning efficiency of dynamic window method.Lastly,the simulation results show that the improvedplanning algorithm is effective and real-time in autonomous obstacle avoidance of mobile robot.Finally,in the built indoor experimental environment,the obstacle position is set randomly,and the improved algorithm is introduced into the ROS system to run the turnlebot3 robot,so that it can walk through and avoid obstacles independently,so as to test the actual effect of the algorithm in this paper.Experimental data show that the improved algorithm has better performance than the original algorithm in path length,path smoothness and tracking time.The research results effectively solve the problems related to robot path planning and autonomous obstacle avoidance,which is beneficial to the development of robot autonomous navigation field.