Research On Environment Modeling And Path Planning For AGV Navigation

With the rapid development of intelligent manufacturing industry,the autonomous and intelligent of automated guided vehicle(AGV)has gradually become the research focus in the field of intelligent logistics.The navigation technology of AGV is one of the keys to ensure its autonomous completion of logistics and transportation tasks.The navigation technology of AGV is a hot research field facing the engineering demand,and also has high academic research value.Therefore,in order to improve the autonomy and intelligence of AGV,this paper studies the map modeling and path planning for the needs of AGV navigation.The main research contents are as follows:A method of map modeling based on feature extraction is proposed.Through clustering and segmentation of the collected point cloud data,the segment fitting and merging of the point cloud in each region after segmentation are carried out,and the required feature segments are extracted.Finally,the indoor map model with obvious structural characteristics is established.On this basis,the vehicle landmark measurement model is established,and Monte Carlo positioning algorithm based on Bayesian posterior probability estimation is used to effectively correlate the odometer state value with the sensor observation value.A path planning method for efficient and dynamic obstacle avoidance is proposed.In order to solve the problem that there are many inflexion points and the inflexion points are not smooth in the path planning results of astar algorithm,Floyd algorithm is used to improve astar algorithm.By removing redundant inflexion points and adding cubic B-spline fitting path,the path is smoother and safer,and the efficiency of path planning is improved.In order to avoid the problem that the traditional dynamic window algorithm is easy to fall into the local optimal value,the vector field histogram algorithm is introduced into the dynamic window algorithm,and the local path is updated by sampling the speed and taking it into the cost function,so as to effectively overcome the local optimal problem in the DWA algorithm.The simulation results show that this method can effectively improve the efficiency of path planning on the basis of dynamic obstacle avoidance.The experimental platform of map modeling and path planning is built.Based on AGV and lidar,a map modeling experimental platform is built.The experimental results show that the proposed method can effectively improve the accuracy of the model.The global path planning experiment of the improved astar algorithm based on Floyd is carried out.The experimental results show that the improved algorithm can effectively improve the speed of path search.At the same time,the experimental results show that theimproved DWA algorithm can effectively avoid dynamic obstacles and local extremum problems.