Research On Autonomous Localization Of The Mobile Robot

With the rapid development of intelligent manufacturing and artificial intelligence,robot technology has made great progress in recent years.Nowadays,robots are playing an increasingly important role in our daily life,especially mobile robots with strong mobility.They are widely used in industrial production,family services,health care,disaster relief and rescue,military and other fields.Therefore,mobile robot technology has gradually become one of the hot research directions.The most important problem of mobile robot in the actual working process is to achieve autonomous positioning and navigation,and the premise of navigation is positioning first.Only when the mobile robot "knows where it is" in the process of working can the mobile robot serve us better.In order to solve this problem,SLAM(Simultaneous Localization and Mapping)technology,which is widely used in unmanned driving,virtual reality,intelligent robots and other fields,is used to study the location of mobile robots.The specific research contents are as follows:Firstly,the principle of SLAM technology is introduced.It includes the mathematical description of SLAM,the model description of SLAM and the map representation method of SLAM technology.It also introduces the current popular laser SLAM and visual SLAM algorithms,and compares their advantages and disadvantages.Secondly,the laser SLAM algorithm based on lidar is analyzed in detail.Including the working principle of lidar,the principle of Monte Carlo localization algorithm in laser SLAM,loop detection and the overall working principle of laser SLAM,and then using laser SLAM to carry out a simple experimental verification;Third,the ORB-SLAM2 algorithm based on visual SLAM is analyzed in detail.Including the principle of ORB-SLAM2 algorithm,the principle of small aperture imaging in camera and camera calibration,and then using this algorithm to obtain the robot's trajectory in the laboratory.Finally,experiments are carried out to verify the actual work of mobile robots.Experiment 1: Aiming at the high positioning accuracy and no cumulative error of thelaser SLAM algorithm,the trajectory of the laser SLAM algorithm is used as the real trajectory,and then the single-eye ORB-SLAM2 algorithm is used to verify the positioning of the mobile robot.The results show that the monocular ORB-SLAM2 algorithm can locate the robot robustly.The second experiment simulates the location of the mobile robot in special environment,and studies the location of the robot with sufficient light and auxiliary lighting equipment respectively.Experiments show that the monocular ORB-SLAM2 algorithm can be applied to the location of mobile robots in special environments under dark conditions with the aid of auxiliary lighting equipment.