Research On Map Construction And Path Planning Of Mobile Robot Based On SLAM

With the rapid development of computer technology,embedded system,artificial intelligence and other fields of science and technology,mobile robots are playing an increasingly important role in the production and life of human beings due to their characteristics of environment awareness,path planning,motion control and cooperative movement.However,the positioning problem is the premise for the application of mobile robots in reality.Traditional robot positioning is completed on the basis of known environmental information,but it is often difficult or even impossible in the actual environment.Therefore,synchronous positioning and map construction in the unknown environment has become a core technology in the field of robot research.For the current intelligent mobile robot Based on SLAM technology,the key technologies of map construction and path planning for mobile robots are studied in this paper.The main contents of the thesis are as follows:Firstly,the hardware and software integrated system of mobile robot which can realize autonomous positioning and path planning is studied and built,and the kinematics model,track deduction model,environment map model,depth camera model,lidar ranging model and odometer model of robot are deduced and proved in detail.Secondly,the mathematical description of SLAM problem is given.Aiming at the problem that a single encoder is likely to produce a large positioning error,the IMU and odeometer data are fused with the extended kalman filter algorithm to improve the positioning accuracy of the robot in real time.To solve the problem that 2D lidar cannot detect obstacles above or below its detection plane in complex 3d environment,the method of depth camera and lidar data fusion is adopted to accurately perceive obstacles at different heights.Three kinds of laser SLAM algorithms are analyzed and compared,and the principle of Cartographer algorithm is derived in detail.Aiming at the problem of its own lag in pose estimation,the framework of Cartographer algorithm is optimized by using the method of integrating odeometer data and IMU data to estimate the pose.Finally,the paper introduces the navigation framework Navigation Stack mobile robot,studies the mainstream adaptive Monte Carlo localization(AMCL)algorithm,analyzes and compares various path planning algorithms,and puts forward an optimization A*algorithm for the disadvantages of high collision rate many times of turning and low smoothness of the traditional A*algorithm,and gives the anti-collision strategy and the path smoothness optimization strategy.The Matlab?Rviz and Stage simulation platform carries on the simulation experiment to the localization and the path planning algorithm,and analyzes the experimental results.