Fusing Time-of-Flight Cameras and Inertial Measurement Units for Ego-Motion Estimation

This paper presents an approach for ego-motion estimation based on a Time-of-Flight (ToF) camera. It is applied in real-time on a mobile robot platform during mission. The proposed method utilizes the coherence of depth and reflectance data of ToF cameras by detecting image features on reflectance data and estimating the motion on related depth data. This motion estimate is fused with data from an inertial measurement unit in order to gain higher accuracy and robustness, especially in situations when image registration fails. Results are benchmarked against reference poses from an accurate laser ranger finder-based localization.