Analysis of a mobile robot equipped with 360 LiDAR utilising gmapping SLAM in indoor and outdoor environment

Akharuddin, Norzalina (2021) Analysis of a mobile robot equipped with 360 LiDAR utilising gmapping SLAM in indoor and outdoor environment. Project Report. Universiti Teknikal Malaysia Melaka, Melaka, Malaysia. (Submitted)

Text (Full Text) Analysis of a mobile robot equipped with 360 LiDAR utilising gmapping SLAM in indoor and outdoor environment.pdf - Submitted Version Download (6MB)

Abstract

Autonomous vehicles have been highly in demand due to the prediction of road accidents based on Royal Malaysia statistics in the year 2020 by the Malaysian Institute of Road Safety Research (MIROS). Based on the ARIMA model produced, it is expected that the number of fatalities will increase to 8,760 (year 2015) and 10,716 (year 2020). More efforts and effective interventions need to be created creatively to prevent fatal road crashes, and to bring down fatalities. This thesis focuses on developing a mobile robot that would replicate the actual concept in order to provide a general idea of its market demands and capabilities by incorporating LiDAR as an additional sensor for prevention to help reduce one of the factors that can cause incidents. The objective is to implement the gmapping SLAM algorithm for the mapping of a mobile robot in the indoor and outdoor environment using the Robotic Operating Systems (ROS). ROS is the major system used for interpreting the mapping data from the laser scan through ROS Visualization (Rviz). The development of a mobile robot from scratch consists of a Jetson Nano, Arduino Nano, RP-LiDAR, 12V Lithium-ion battery, 12V DC motor with encoder, switch, and an LM2596 DC-DC adjustable step-down converter. The mobile robot is integrated with ROS and the mapping procedure occurs in an indoor and outdoor environment. The 2D map produced by the RP-LiDAR A1 scanner was evaluated in terms of accuracy, dependability, and the laser scanner’s limitations for mapping in specific environments.

Actions (login required)

View Item