Host: The Japan Society of Mechanical Engineers
Name : [in Japanese]
Date : May 10, 2017 - May 13, 2017
The paper proposes a stability improvement method of 6-DoF MCL (Monte-Calro Localization) for arbitrary motion of hand-held sensing application. In previous work[3], odometry-free 6-DoF localization was achieved using 6-axis IMU and 3D-LIDAR with the designed inertial state vector. Based on this method, we first examine the effect of each state with basic sensor motions. Then, we propose two modification; 1) point cloud extraction for better shape matching, 2) separated calculation of position and orientation. The basic evaluation results show that the proposed method improves the stability for known issues; at yawing motion and motion with tilted sensor.
