Design And Implementation Of Robot Vision Navigation System Based On ROS

Vision-based self-localization and navigation technology of mobile robots is a new hot research topic in recent years,which is convenient for our life and work in many aspects.This paper is based on the ROS operating system of Ubuntu,which combines the navigation of monocular camera with the navigation of IMU module to realize the autonomous navigation and positioning of mobile robot.There are scale uncertainties in monocular vision navigation,and cumulative errors in inertial navigation for a long time.Combining the advantages and disadvantages of the two methods,the high precision characteristics of monocular camera navigation and the stability of IMU module navigation are combined to improve the navigation accuracy and reliability.Specific research contents are as follows:This paper studies the basic principle of IMU,chooses strapdown navigation system as inertial navigation platform and introduces relevant knowledge,introduces the components and software architecture of inertial navigation under ROS operating system,calibrates IMU module and carries out positioning experiments on intelligent car under ROS operating system to analyze the navigation accuracy of IMU navigation.The basic principle of monocular vision navigation is studied.The monocular camera is calibrated by Zhang's plane method.The visual information collected by the projectile screen camera is processed by SFM algorithm combined with BA joint optimization method to navigate the intelligent car.The navigation accuracy of the monocular camera is verified by experiments.Based on the navigation errors of MU and monocular camera respectively,a VIO algorithm of IMU fusion monocular vision is proposed.The coordinate errors between them are eliminated by initialization of monocular camera and joint initialization of inertial module monocular camera.BA joint optimization is used to optimize the collected information,and back-end optimization is added to eliminate the global error.The positioning and navigation of intelligent robots are realized.The feasibility of VIO algorithm based on IMU fusion monocular vision is proved to be stable under the intelligent car.