Visual SLAM based on EKF filtering algorithm from omnidirectional camera

Abstract

The SLAM(simultaneous localization and mapping) problem is one of the essential challenges in mobile robotics. In this paper, we integrate a method to solve the visual SLAM problem based on the extended Kalman filter (EKF) algorithm with an omnidirectional camera. The features in environment around the mobile sensor are extracted by a high-speed corner detection method using omnidirectional vision. We use the spherical camera to get the geometric information from the image sequences. Due to the large field of view, we can obtain robust estimates. The simulation result indicates that the EKF method used applied to spherical camera is effectiveness.