Visual Mapping And Path Planning Of Mobile Robots In Indoor Scenes

With the rapid development of robot technology and intelligent control technology,the mapping and path planning of mobile robots have been widely used in industrial production,smart home and other fields,and have gradually become an important part of people's production and life.In the field of mobile robot mapping,the visual sensors have the advantages of low cost and high scene recognition ability,and have become the main way to obtain environmental information.At the same time,with the gradual expansion of the demand for mobile robots,people have higher requirements for the intelligence of mobile robots,so in-depth research is needed.This thesis takes the visual navigation of mobile robots in indoor scenes as the application background,and based on vision and robot operating system to carry out related research on the mapping and path planning of mobile robots.The main work is as follows:Construction of visual navigation system for omnidirectional mobile robot.The hardware platform is built with small omnidirectional mobile robot,and the software platform is built with Linux operating system and robot operating system.The coordinate system between the mobile robot and the depth camera is established,and the internal and external parameters and depth alignment of the camera are calibrated.At the same time,the kinematics model of mecanum wheel is solved to realize omnidirectional control of the mobile robot.Based on the Simultaneous Localization and Mapping(SLAM)and target detection algorithm,a mobile robot map construction and positioning method is designed.First,visual map construction algorithm is used to establish environment sparse point cloud maps,and through the point cloud processing method is transformed into a dense point cloud and octree map,and then the projection to plane to generate the raster map used in mobile robot navigation.At the same time,the target detection algorithm is used to identify the environmental landmarks and map their information to the map.When the mobile robot navigates through the map,the difference between the current position of the road sign and the historical position is used to correct the position of the mobile robot and improve its positioning stability.Based on global path planning algorithm and graphics algorithm,a curved path smoothing method for mobile robot is designed.First,the global path planning algorithm is used to plan the initial global path,and then the large turn of the path is judged.Then the Bresenham circle drawing algorithm is applied to optimize the path on the basis of the initial path,so as to improve the speed and stability of the mobile robot in the curve.Experimental verification is carried out based on the omnidirectional mobile robot visual navigation system.In the data set and specific experimental environment,the design of the mobile robot map construction and positioning method,path optimization algorithm is verified by experiments,the experimental results prove the effectiveness and feasibility of the design method.