AUTONOMOUS VEHICLE MAPPING WITHVLP-16 LIDAR USING LIDAR ODOMETRY AND MAPPING ALGORITHM

Abstract

Mapping is one of the important aspects in Simultaneous Localization and Mapping (SLAM) methods. It is imperative for an autonomous vehicle (AV) to be able to map its surrounding environment in an accurate manner so that it can be used by the other systems in the AV. This led to the issue thatfaced by many SLAM researches that asks for the possibility for an AV to simultaneously build a consistent map of the surrounding unknown environment and verify its location within the said map. One of the obstacles in mapping the AV is localization estimation. Scattered GPS signal problem is a known factor held by the earlier AVs and to counter this issue Light Detection and Ranging (LiDAR) sensor was utilised due to its ability to collect large amount of data in a short amount of time. This research aims to provide the AV mapping by using LIDAR focusing on the optimization-based method. Data was collected using a 3D LiDAR sensor, Velodyne Puck in the selected area around UniversitiTeknologi MARA. This project was experimented with Robotic Operating System (ROS) and visualized through the RViz tool in ROS. This research employed Lidar Odometry and Mapping (LOAM) algorithm in building the map and has produced a viable result with less than 5% percent error rate. It demonstrates the benefits of the LOAM algorithm on the real-world data. With the high accuracy achieved shows this algorithm can help in the development of autonomous vehicle for the mapping process. Future works can be done to ensure the performance of the system is in excellent condition and increase its accuracy.