Pose estimation with lidar odometry and cellular pseudoranges

A pose estimation framework by fusing light detection and ranging (lidar) odometry measurements and cellular pseudoranges using an extended Kalman filter is proposed. Iterative closest point (ICP) is used to solve for the relative pose between lidar scans. A maximum likelihood estimator is developed for lidar scan registration. The proposed framework works with few ICP iterations; hence, can be used for real-time applications. The framework is tested experimentally, and it is demonstrated that the two-dimensional position root mean square error obtained with ICP only can be reduced by 93.58% by fusing lidar odometry and cellular pseudoranges.