Position and Orientation Control of a Mobile Robot based on Robust Marker Position Estimation using a Sphere Camera

Abstract

- This paper describes position and orientation control of a mobile robot using a sphere camera. While a sphere camera has a wider field of view than a normal camera, the field of view is distorted. Therefore, image recognition is performed by compensating for this distortion. The control of a mobile robot using this sphere camera enables better localization estimation, distance recognition, and object recognition than with a normal camera. In this paper, we propose a position estimation method using markers with color invariants that are robust to illumination. We use a hybrid marker system that employs a circle marker and an AR marker. We demonstrate the effectiveness of our method by conducting experiments on position control of a two-wheel mobile robot using the sphere camera and marker recognition.