Relative pose measurement of moving rigid bodies based on binocular vision

Abstract

To solve the shortcomings of the traditional monocular vision measurement algorithm, which needs to be solved iteratively, this paper proposes a method based on binocular vision for measuring the relative pose of moving rigid bodies. In this paper, the coordinates of the feature points of the follower rigid body coordinate system are obtained by using the method of least squares. At the same time, combined with the dual quaternion algorithm, the leader and follower rigid bodies’ coordinate systems are uniformly described, and the model takes the minimum sum of squared errors as the objective function and solves the pose parameters by the dual quaternion method. The simulation results show that the dual quaternion algorithm not only overcomes the shortcomings of the traditional quaternion and translation vector split description coordinate system transformation but also has strong stability; the relative position error is less than 0.05 m, and the relative attitude error is less than 0.2°, which can meet the measurement accuracy requirements of the relative pose of the moving rigid body.