Processing of Lidar and IMU Data for Target Detection and Odometry of a Mobile Robot

Abstract

In this paper, the processing of the data of a 3D light detection and distance measurement (LiDAR) sensor mounted on a mobile robot is shown, introducing an innovative methodology to manage the data and extracting useful information. The LiDAR sensor is placed on a mobile robot with a modular design that permits the easy change of the number of wheels, which was designed to travel through several environments, saving energy by preliminarily changing the number and arrangement of the wheels in the environment. Furthermore, the robot can recognise landmark in a structured environment by using a classification technique on each frame acquired by the LiDAR. Furthermore, considering the experimental tests, a new simple algorithm based on the LiDAR data processing together with the inertial data (IMU sensor) through a Kalman filter is proposed to characterise the robot’s pose by surrounding environment with fixed landmarks. Finally, the limits of the proposed algorithm have been analysed, highlighting new improvements in the future perspective development for permitting autonomous navigation and environment perception with a simple, modular, and low-cost device.