Research on robot positioning technology based on inertial system and vision system

Aiming at the problem that the positioning accuracy of visual odometer decreases due to the ambiguity of inertial motion in the process of robot high-speed motion, a research method of robot positioning technology based on inertial system and vision system is proposed. Using extended Kalman filter and inertial sensor heading estimation, the observation equation is established with visual odometer as relative linear velocity and relative angular velocity sensor. Considering the plane motion of the robot, there is no lateral and longitudinal jump and slip. The robot positioning method is established under the constraint of zero instantaneous velocity, which overcomes the shortcomings of visual positioning and inertial positioning, and can effectively improve the positioning accuracy. The experimental results show that the effect is better than the traditional method in the practical application.