Localization Fault Detection Method using 2D LiDAR and Fisheye Camera for an Autonomous Mobile Robot Control

Localization is an important function of autonomous mobile robots, and autonomous actions caused by incorrectly estimated robot posture may lead to major accidents. In this study, we propose a fault detection function of localization using 2D LiDAR and fisheye camera. The proposed fault detection method focuses on the likelihood to detect faults, which can be improved by suppressing the update of the particle filter based localization. We conducted several experiments in real environments to verify the effectiveness of our proposed method.