Planar region alignment based ego-motion estimation for mobile robot

Abstract

The ego-motion estimation is a basic task in many operations of a mobile robot. An accurate and robust result of motion estimation can greatly improve the operation. In this paper, we assume the mobile robot with visual observer is moved on flat ground and a constrained camera-robot configuration is used to simplify the camera motion model. Based on the assumption and simplified camera-robot configuration, the 3D camera motion can be separately estimated by planar region alignment between two frames. Firstly, a detected 2D motion between two frames is used to align corresponding planar regions, such an alignment or registration removes the effects of camera rotation, and the resulting residual parallax displacement field between the two region-aligned images is an epipolar field centered at the FOE (Focus-of-Expansion). Then the camera translation can be recovered from the epipolar field. The camera rotation is recovered from the computed 3D translation and the detected 2D motion. The camera motion model is derived and expressed by two parts, the multistage process for ego-motion estimation is introduced. Some experiments with real images demonstrate the robustness of the motion estimation. And the factors which may affect the results are discussed.