Research On Path Planning Algorithm Of Mobile Robot In Intelligent Warehouse

The rapid development of e-commerce industry promotes the transformation of logistics industry from labor-intensive to technology intensive.Intelligent logistics has become the development trend of logistics industry,followed by the application and popularization of intelligent logistics warehouse and various advanced supporting technical equipment.The intelligent mobile robot with multi-function of handling,palletizing and sorting has become one of the indispensable intelligent equipment in the intelligent logistics warehouse.The path planning problem is the key and primary problem needed to be solved for widely use of many mobile intelligence agents,such as the intelligent mobile robot.Firstly,this paper proposes a path planning algorithm for mobile robot based on visibility points(Visibility Points Searching Algorithm,VP-S),mainly including visibility point selection and path planning.The visibility point selection preprocesses the obstacle vertices in the environment,and gets the set of visibility points from the current position of the mobile robot,it narrows the search scope of the algorithm.The path planning searches the optimal node of the best path from the set of visibility points,and plans an obstacle-free path from the start point to the end point for the mobile robot by using heuristic function.Although VP-S algorithm can plan an obstacle-free path,it has some defects such as insufficient security,many redundant nodes and low smoothness,besides it only considers the single task of mobile robot,not suitable for the multi task environment in intelligent logistics warehouse.To fix the defects of VP-S algorithm,the algorithm is improved in four aspects one by one: security,redundant nodes,smoothness,dynamic obstacle avoidance and multi goals.In order to solve the lack of security of the path planned by VP-S algorithm,a safe path planning algorithm based on visibility points is designed.A virtual volume expansion parameter enlarge?index is added to the algorithm to guarantee the safety of the path planned by the algorithm;Aiming at fix the problem that there are many redundant nodes in the path planned by VP-S algorithm,a path planning algorithm without redundant nodes based on visibility points is proposed.It calculates and filters the redundant nodes of the path from the starting point until to the final point,retains the key nodes that the path must pass.It makes sure there are fewer turning points and the length of the path is shorter.In order to solve the problem that the smoothness of the path planned by VP-S algorithm is low and does not conform to the motion characteristics of mobile robot,a smooth path planning algorithm based on visibility points is designed.The dynamic window algorithm is improved by adding the guidance of the line connecting the starting point and the end point to the heuristic function.And the course angle deviation threshold is also set to correct the course angle of the mobile robot in real time.Then the improved dynamic window algorithm is integrated with the VP-S algorithm to improve the smoothness and realize dynamic obstacle avoidance,and also make VP-S algorithm have the ability of dynamic local planning.Next,considering the multi task of the intelligent warehouse,a multi-objective path planning algorithm based on visibility points is proposed,which enables the mobile robot to reach multiple targets and complete multiple tasks in a single planning process.Finally,the simulation in the Robot Operation System(ROS)verifies the feasibility and optimization of the VP-S algorithm and the improved VP-S algorithm in the intelligent logistics warehouse and practical scene.The comparison simulation experiments in different complexity and different size scenarios with the Ant Colony Optimization,A* algorithm and other path planning algorithms verify the optimization of the length,security and smoothness of the path of VP-S algorithm and the improved VP-S algorithm.