Application of IMU Pre-integration in Variable-Height Lidar Odometry

Abstract

In this paper, we propose a variable-height lidar odometry method to enhance the accuracy of the mobile robot motion estimation for an indoor environment. The lidar device adopted a novel variable-height lidar, IMU pre-integration method, and a matching algorithm with line and surface feature matching is presented. With the new algorithm, the edge points and plane points from the point clouds are employed to fit line segments and planes. Based on the distance between matching features and IMU pre-integration information to build a nonlinear optimal estimation equation. We compare the performance of our method with the wheel encoder odometry method. Experiment results show that the proposed multi-sensor fusion method has high pose precision.