2D Visual Odometry method for Global Positioning Measurement

Abstract

The goal of this paper is to develop a method for estimating the 2D trajectory of a road vehicle using visual odometry. To do so, the ego-motion of the vehicle relative to the road is computed using a stereo-vision system mounted next to the rear view mirror. Feature points are computed using Harris detector. After that, features are matched between pairs of frames and linked into 2D trajectories. A photogrametric approach is proposed to solve the non-linear equations using a least-squared approximation. The purpose is to merge trajectory information provided by the visual odometry system with information provided by other sensors, such as GPS, in order to produce really accurate measurements of vehicle position. Providing assistance to drivers is among the prime applications of the proposed method. Nonetheless, other applications such as autonomous robot or vehicle navigation are also considered. The proposed method has been tested in real traffic conditions without using prior knowledge about the scene nor the vehicle motion. We provide examples of estimated vehicle trajectories using the proposed method and discuss the key issues for further improvement.