Research On Mobile Robot Location Based On Inertial Navigation And Visual Information Fusionbased On ROS

The positioning of mobile robot is the basis of autonomous navigation,and it is also an important research topic in mobile robot research field.The main research content of this paper is based on ROS operating system,the inertial navigation and visual sensor information fusion,to solve the problem of mobile robot autonomous positioning.At present,monotonous SLAM has initialized scale problem and tracking scale drift problem;inertial navigation data is stable,but accumulation error is more serious.The fusion of these two technologies can combine the high accuracy of monocular vision SLAM with the stability of inertial navigation data,and learn from each other to achieve the purpose of positioning accuracy.The specific research contents are as follows:Firstly,the basic positioning theory of strapdown inertial navigation system is studied,including inertial navigation coordinate system,basic rotation theory and knowledge of attitude solution based on quaternion algorithm and the theory of dead reckoning.Lay the foundation for further research and application;Secondly,the theoretical basis of visual positioning of mobile robot is studied.Firstly,the relationship between camera imaging and four coordinate systems is introduced.Create a camera imaging model to calibrate the camera parameters.On this basis,a mobile robot vision positioning technology based on POSIT algorithm is proposed;Third,because the inertial navigation and monocular SLAM are local localization methods,the global information is missing and the cumulative error can not be corrected.In this paper,we study the method of visual simultaneous positioning and terrain construction based on inertial navigation,and propose a visual inertial SLAM algorithm based on ORB_SLAM system to integrate IMU information;Fourth,the hardware and software systems and experimental parts of the entire platform are introduced.Firstly,the hardware and software platform used in the experiment are introduced,and the motion model of the two-wheel differential mobile robot is established.The ROS operating system and its common functions,as well as the basics of navigation and positioning,are briefly described.The analysis and design of the experiment takes into account the contrast of many different situations.Provides an effective,error and positioning accuracy analysis method.Experimental analysis shows that the combination of inertial navigation and monocular SLAM method effectively improves the positioning accuracy,and shows the practical application of the algorithm.