LIWO-SLAM: A LiDAR, IMU, and Wheel Odometry Simultaneous Localization and Mapping System for GNSS-Denied Environments Based on Factor Graph Optimization

Abstract

The paper presents a sensor fusion algorithm for Simultaneous Localization and Mapping (SLAM) for a wheeled robot which fuses LiDAR, IMU, and wheel odometry using a factor graph. Building on the existing algorithm LIO-SAM, a detailed derivation is given for including an odometry factor for a four-wheel-independent steering and four-wheel-independent driving (4WIS4WID) robot. The algorithm is tested and evaluated using datasets collected at a tunnel research facility. The results show that in comparison to an Extended Kalman Filter, drift can be significantly reduced.