Research On Autonomous Exploration And Real-time Obstacle Avoidance Method Of Robot Based On ROS

The simultaneous localization and map building of mobile robots are the preconditions of autonomous navigation technology.It provides scene information for mobile robots to make next control strategy and improves the decision-making efficiency of mobile robots.Autonomous exploration of mobile robots can be interpreted as the process of simultaneous localization and map building(SLAM)autonomously by mobile robots.It is a research hotspot in the field of autonomous navigation of mobile robots.Real-time obstacle avoidance technology of mobile robots is the key in the field of autonomous navigation technology of mobile robots and determines the success or failure of autonomous navigation of mobile robots.mobile robots need to be more autonomous and fast in the process of simultaneous localization and map building in closed scenes,reduce labor costs and improve economic benefits.In the face of more and more vehicles and pedestrians,traditional obstacle avoidance methods of mobile robots can no longer cope with increasingly complex traffic conditions.Researchers began to do in-depth research on improving obstacle avoidance efficiency of mobile robots.The obstacle avoidance research of mobile robots not only considers the,driving reliability.but also considers the driving safety and speediness of vehicles.Using advanced algorithm to design obstacle avoidance of mobile robots can effectively alleviate road congestion,reduce the probability of traffic accidents and improve driving safety.In this thesis,three aspects of the existing autonomous navigation technology are studied.The main research work of this thesis is as follows:(1)Aiming at the problem of simultaneous localization and map building,this paper uses two-dimensional lidar to collect scene information and improve the accuracy and real-time of scene map building on the basis of original SLAM.In the process of scanning and matching of scene map building,this paper combines gradient descent method with neighboring region method,To some extent,theefficiency of SLAM and the accuracy of map construction are improved by building scene maps in real time at different speed of mobile robots.(2)Aiming at the problem of autonomous exploration of mobile robots,this paper pays special attention to the problem of map integrity and timeliness,we use the method of Frontier target points to explore the target points,we divide the target points into local target points and global target points by fast expanding random numbers,and determines the exploration target points by combining the index function of information gain,which improves the real-time and map integrity of autonomous exploration.Finally,A* algorithm is used to plan the path of mobile robot,which improves the efficiency of mobile robot exploration.(3)Aiming at the problem of obstacle avoidance of mobile robots,this thesis fully considers the real-time obstacle avoidance of mobile robots with scene map information.Firstly,obstacles in the process of mobile robots' movement are divided into static obstacles and dynamic obstacles.When avoiding static obstacles,on the basis of vector field histogram method,an adaptive threshold avoidance method is proposed to solve the problem of threshold sensitivity and dead zone of mobile robot.When avoiding dynamic obstacles,an improved relative coordinate system obstacle avoidance method is proposed.In the process of obstacle avoidance,the influence of the size of mobile robot on obstacle avoidance effect is considered.Finally,the effectiveness of the proposed method is verified by simulations and experiments.