Intelligent Transfer Vehicles For Raw Grain Storage Path Planning Studies

With the bumper harvest of grain year after year,the function of grain depot as the core unit of grain storage is very important.During the grain harvest season,raw grain storage faces the problem of inefficiency.In order to improve the grain transfer efficiency of grain depots,this paper takes the intelligent transfer vehicle of grain depot as the development object and studies the path planning algorithm applied to intelligent transfer vehicle.In this paper,the environmental modeling and map positioning methods are studied,the kinematics model of the vehicle is established,and the improvement and optimization of the global path planning algorithm,local path planning algorithm and path planning algorithm of intelligent transfer vehicle are studied.The autonomous path planning function of intelligent transfer vehicle is realized.Firstly,by studying different environmental mapping methods,a raster method suitable for modeling the grain depot environment is selected for the special environment of the grain depot,and a method for drawing a raster map using high-precision latitude and longitude coordinates is proposed.Through the study of the kinematic models of Ackerman model and bicycle model,and the characteristics of the two kinematic models are compared and analyzed,the cyclical kinematics model suitable for intelligent transfer carts is selected.The accurate position of the vehicle in the map is obtained through the relevant coordinate conversion of high-precision latitude and longitude coordinates.Vehicle mileage is calculated by changing the map position in real time.The effect of coordinate conversion and mileage calculation is tested by Rviz of the real vehicle,and the deviation between the position and the actual position after the coordinate conversion is compared and analyzed.The experimental results show that the position accuracy,coordinate conversion and mileage calculation effect after coordinate conversion meet the requirements of intelligent transfer carts.Then,the global path planning algorithm using raster map is studied,and several global path planning algorithms are simulated and analyzed,and the simulation results show that compared with other algorithms,the A* algorithm has advantages in searching nodes,search time,path turning points and finding target points,and the A* algorithm is selected as the global path planning algorithm of intelligent transfer cart according to the simulation results.In this paper,two local path planning algorithms,TEB algorithm and artificial potential field algorithm,are studied,and the two algorithms are simulated and tested.The simulation results show that compared with the artificial potential field algorithm,the TEB algorithm has the advantages of obstacle avoidance path smoothness,closest distance to obstacles and vehicle motion control,and the TEB algorithm is selected as the local path planning algorithm based on the simulation results.Secondly,this paper optimizes the path planning algorithm.In view of the problems of the traditional A* algorithm searching for paths in grain warehouses,the expansion node traverses obstacles for many times,resulting in a decrease in search efficiency,the planned path is close to the obstacle and there is a risk of collision,and the lack of kinematic constraints is not conducive to vehicle motion control,this paper improves the A* algorithm as follows:the obstacles are processed by fusing the DBSCAN clustering algorithm and the K-means clustering algorithm,the obstacles are clustered and the clustering center point and boundary point of each type of obstacle are extracted,and the boundary point is used instead of the obstacle.Improved the efficiency of the algorithm.The operation efficiency of A* algorithm after boundary points replace obstacles is simulated and analyzed,and the simulation comparison results show that the efficiency of A* algorithm search path is greatly improved.The domain expansion mode of A* algorithm is optimized to expand the scope of domain expansion.In order to make the improved domain expansion method applicable to the obstacle boundary point raster map,a node abandonment strategy is added for the improved domain expansion method.In order to keep the planned path away from obstacles,add obstacle avoidance strategies for the improved way of expanding the domain.Compared with the traditional domain expansion mode,the domain expansion mode of the improved domain expansion method of the node abandonment strategy and the domain expansion mode of the improved obstacle avoidance strategy,the A* algorithm of the traditional domain expansion mode,the domain expansion mode of the improved node abandonment strategy and the improved obstacle avoidance strategy of the fusion node are analyzed,and the simulation results show that the improved domain expansion mode of the fusion node obstacle avoidance strategy and the improved domain expansion method of the fusion node abandonment strategy have improved algorithm efficiency and enhanced obstacle avoidance performance.By optimizing the algorithm heuristic function,the repulsive field function of the artificial potential field algorithm is added to the A* heuristic function as a safety function,and the artificial potential field algorithm repulsive field is introduced for the clustering center point and some boundary points of the obstacle,so as to increase the obstacle avoidance ability of the A* algorithm and avoid the planned route from entering the high-density obstacle area.The simulation comparison analysis results show that the heuristic function added to the artificial potential field algorithm has the ability to avoid high-density areas and maintain a certain distance from obstacles.By using TEB algorithm mixed with A* algorithm for path planning,kinematic constraints are added to the movement of the vehicle,and the handling and obstacle avoidance ability of the vehicle are improved.The simulation results show that the TEB algorithm has the ability to follow the global path point and obstacle avoidance,and the path planned by the mixed path planning algorithm is smooth,which is suitable for the transfer needs of intelligent transfer vehicles in grain storage roads.Finally,this paper conducts real vehicle experiments and builds corresponding experimental platforms.Using the move＿base function package,the autonomous navigation framework of intelligent transfer vehicle was constructed,and the experimental site was selected to carry out the effectiveness experiment of the improvement of the heuristic function of the global path planning algorithm of the intelligent transfer vehicle and the feasibility of A* algorithm hybrid TEB algorithm path planning.In a simulated grain warehouse environment,the experiment tested the vehicle’s ability to track the navigation point to transport grain.Experimental test results show the effectiveness of the heuristic function improvement of the global path programming A* algorithm,and the scheme of mixed path programming is also very feasible.Intelligent transfer vehicles can be used for grain transfer operations in grain warehouses.