Vision-based controller for autonomous UGV

• Main Author: Singaram, Venkatachalam
• Other Authors: Wang Jianliang
• Format: Final Year Project
• Language: English
• Published: 2017
• Subjects: DRNTU::Engineering::Electrical and electronic engineering
• Online Access: http://hdl.handle.net/10356/70848
• Institution: Nanyang Technological University

In recent times, the use of mobile robots has significantly increased in different sectors of the industry. Mobile Robots can help us achieve tasks that are physically challenging, harmful, repetitive or inaccessible for human beings. Such systems can be used to accomplish tasks such as inspection and maintenance of buildings, paint jobs of difficult corners and ceilings, transportation of goods and even in agriculture. This project consists of the hardware and software components for an unmanned ground vehicle (UGV) with holonomic movement capability and autonomous movement capability in an indoor environment. The holonomic movement gives the robot the advantage to move the robot in tight spaces maintaining its yaw orientation. For the hardware component of this project, the focus was on assembling the chassis of the UGV and fabricating the support structure for different modules of the robot. The software component of this project emphasizes on writing the low-level firmware for the robot as well as interfacing with Robot Operating System (ROS) for developing high level applications such as path planning and autonomous navigation using Simultaneous Localization and Mapping (SLAM). With this technique, the robot will be able to successfully navigate autonomously in an indoor environment. Additionally, to enhance the application scope of the project a robotic arm is planned to be mounted on top of the robot. Hence, a 6DOF low level arm controller board was designed and fabricated. There are seven Teensy 3.2 controllers which works on I2C master-slave concept. Further, this project can be enhanced by integrating a robotic arm to the UGV and making it as one inter-connected system. This can be achieved by modelling the system and interfacing the robotic arm for high level applications by using ROS packages such as MoveIt!, etc.