LAUSR

This study presents a visualization system in a third-person view for the teleoperation of a snake-like robot. The proposed system visualizes the robot, its surroundings, past path, and predicted path. The visualization is achieved by drawing a robot at self-position and the sensor information, such as a point cloud around the robot, in a virtual 3D space. Therefore, localization plays a key role in visualization. We propose a new localization method that relies only on the information from joint angles and an IMU. We evaluated the proposed method by comparing it with other methods on a flat ground and on a combination of a plane and a slope with a rough surface. The results demonstrated the effectiveness of the proposed method in both cases. Then, we examined the visualization system in room environments. The results showed that the proposed system successfully visualized the robot and its surroundings. We also confirmed the effectiveness of the visual aid by the predicted path. GRAPHICAL ABSTRACT