Visual odometry based on a Bernoulli filter

Abstract

In this paper, we propose a Bernoulli filter for estimating a vehicle's trajectory under random finite set (RFS) framework. In contrast to other approaches, ego-motion vector is considered as the state of an extended target while the features are considered as multiple measurements that originated from the target. The Bernoulli filter estimates the state of the extended target instead of tracking individual features, which presents a recursive filtering framework in the presence of high association uncertainty. Experimental results illustrate that the proposed approach exhibits good robustness under real traffic scenarios.

Image Gradient-based Monocular Visual-Inertial Odometry

“…An accurate ego-motion estimation of a vehicle in a global navigation satellite system (GNSS) denied environment has been a challenging task in robotics and autonomous driving for several decades [1]. To tackle this issue, the ego-motion estimation algorithm using camera information, so-called Visual Odometry (VO), has been studied [2], [3], [4]. VO incrementally estimates a relative 6-DOF pose between consecutive images.…”