Navigation Control Of Mobile Robot Based On Improved A* Algorithm Of Jump Point Search

The navigation technology of mobile robot is the key research direction in the field of robot.Among the commonly used navigation methods,lidar navigation has been widely used because of its high precision,good real-time performance and strong environmental adaptability.In this paper,two-dimensional lidar is used to realize the navigation system of mobile robot,and three key problems of autonomous positioning,map construction and path planning in the navigation system of mobile robot are studied in detail.Combining with the thought of hop search algorithm,the traditional A*pathfinding algorithm is improved,and the indoor navigation system of mobile robot is realized.The main research contents of this paper are as follows:Firstly,the algorithm implementation principle of GMapping-SLAM framework was introduced,and the advantages of GMapping-SLAM framework in small indoor scenes were analyzed by comparing with Hector-SLAM and Cartograph-SLAM algorithms.Combined with the actual experimental environment in this paper,GMapping-SLAM framework was finally selected as the main framework for positioning and map construction in this paper.In the simulation environment,positioning and map construction were carried out to obtain a complete two-dimensional raster map of the environment and save it for subsequent navigation and positioning calls.Secondly,the traditional A*pathfinding algorithm is improved with the thought of jump point search algorithm.The search strategy of the traditional A*pathfinding algorithm itself will produce many redundant computing nodes,which directly leads to some defects of the traditional A*algorithm in the path planning in large scenarios,such as large computation,heavy memory consumption and slow pathfinding speed.This article will jump point search algorithm ideas into traditional A*algorithm,through defined in raster map key nodes:"jump points",can guarantee to obtain the optimal path at the same time,significantly improve the speed of mobile robot pathfinding,reduce the memory overhead,especially in the larger context of hoisting system of the real-time effect is more obvious.Finally,the performance of the proposed algorithm and navigation system is verified by practical experiments.Firstly,the experiment of location and map construction was carried out in the unknown indoor environment,and the raster map meeting the navigation requirements was obtained through GMapping-SLAM.Secondly,the navigation error of the system is measured as centimeter level through the autonomous navigation experiment of the given target point,which meets the accuracy requirements of the indoor service robot.Then,through the comparative effect of the path planning algorithm,it is verified that the improved A*algorithm can significantly improve the pathfinding speed and enhance the real-time performance.Finally,the dynamic obstacle avoidance test is carried out to confirm that the navigation system can meet the requirements of dynamic obstacle avoidance and ensure the safety of navigation.To sum up,this paper use two-dimensional lidar to realize the navigation system of mobile robot,and makes a detailed study on the three problems of autonomous positioning,map construction and path planning of mobile robot in the navigation system.A set of indoor navigation system with high precision,good security and strong portability is obtained.