This paper describes a land-based navigation system that integrates inertial sensors and odometer with Light Detection and Ranging (LiDAR). In the proposed multisensor navigation system, the popular point-based scan matching algorithm Iterative Closest Point (ICP) is used to estimate the vehicle's relative translational and rotational changes from raw LiDAR measurements without the need for feature extraction. To accelerate the ICP algorithm and improve the accuracy, the outputs of Inertial Navigation System (INS) and odometry are used as an initial motion guess to enhance the scan matching algorithm by reducing the error in initial alignment. The relative translation and rotation changes from LiDAR are fused with changes from INS/Odometry through Extended Kalman Filter (EKF) in a tightly coupled scheme. Real experiments were conducted to evaluate the performance of the proposed system. Results showed that by integrating LiDAR with INS/Odometry, inertial sensors biases are accurately estimated and the error accumulation in navigation solutions derived from INS/Odometry can be significantly reduced. Meanwhile, with inertial aiding, the accuracy and the convergence time of LiDAR scan matching can be improved as well. Finally, a map of the operating environment is generated based on the corrected pose of the vehicle.

Additional Metadata
Conference 27th International Technical Meeting of the Satellite Division of the Institute of Navigation, ION GNSS 2014
Citation
Liu, S. (Shifei), Atia, M, Gao, Y. (Yanbin), Givigi, S. (Sidney), & Noureldin, A. (Aboelmagd). (2014). An inertial-aided LiDAR scan matching algorithm for multisensor land-based navigation. In 27th International Technical Meeting of the Satellite Division of the Institute of Navigation, ION GNSS 2014 (pp. 2089–2096).