Research On Dynamic Path Planning Method Based On Vision AGV

With the rapid development of AGV industry,AGV path planning research has gradually become a hot spot.However,most of the current research on AGV path planning focuses on the static environment,and does not maximize the potential of AGV work.In this paper,visual AGV is taken as the research object,and the path planning method of visual AGV in dynamic environment is discussed.Three typical dynamic environments,i.e.local dynamic obstacle avoidance,mobile single target dynamic environment and multi-target service point dynamic environment,are studied respectively.The combination of simulation and reasoning is used to verify the effectiveness and feasibility of this method,which provides a certain reference value for the development of AGV path planning to deal with complex environment.This paper studies the positioning principle and positioning information solution method of vision AGV with depth camera as sensor;discusses the difference and application environment between local positioning and global positioning,which lays the foundation for the research of path planning;comprehensively analyzes the characteristics of all kinds of maps,and finally obtains that 2D grid map is more conducive to the research of path planning than other maps.Aiming at the problem that the vision AGV is in the unknown map environment,a dynamic path planning algorithm based on the improved DWA is proposed.Firstly,the shortcomings of the traditional local dynamic path planning algorithm and the causes of the algorithm defects are analyzed.Then,the fuzzy control is used to adjust the weight coefficients of the direction,safety and fast driving in real time.Finally,the simulation experiment is carried out with MATLAB Based on the improved DWA algorithm,it can avoid getting into the locked state while avoiding obstacles quickly.The method of dynamic path planning for mobile single target is studied.Aiming at the situation that the position of the target point changes,a dynamic path planning method based on the reduced point and non reduced point is proposed.For reducing point type,the logic reasoning is used to prove the rationality of the dynamic path planning method based on reducing point type proposed in this paper;the dynamic path planning method for non reducing point type mobile target is mainly studied,and the dynamic path planning method combining historical data and real-time data is proposed.Finally,the effective result of the algorithm in the map environment set in this paper is obtained through the simulation experiment power is 77.5%.A dynamic path planning method for multi-objective service point is studied.In view of the change of path planning task in multi service objectives in AGV movement,a dynamic path planning method based on multi-objective service point is proposed,and the shortcomings of traditional path planning methods are analyzed.The path planning task is judged in real time by establishing a path planning evaluation function,the global optimal path is selected continuously,and the dynamic path planning is completed by using genetic algorithm.The simulation results show that the total path length of this method is 33.96 m less than that of the traditional sequential method,and the local optimal service trap is avoided.