Dynamic obstacle avoidance for robot

• Main Author: Koh, Gim Sheng
• Other Authors: Hu Guoqiang
• Format: Final Year Project
• Language: English
• Published: 2014
• Subjects: DRNTU::Engineering; DRNTU::Engineering::Electrical and electronic engineering::Control and instrumentation::Robotics
• Online Access: http://hdl.handle.net/10356/60816
• Institution: Nanyang Technological University

As the field of Robotics is generally young compared to many others, there are many fields in which engineers and researchers alike have not been able to fully explore. One of such topic in Robotics is obstacle avoidance. The project will not only focus on the implementation and testing of obstacle avoidance algorithms, it will also include the development of a cheaper and smaller sized robot for any future research. Current implementation of obstacle avoidance robots within the school are large and bulky, due to the requirement of having a powerful processor to cater to the large amount of processing required. Therefore, the aim of this project is to design and implement an autonomous robotic solution that is compact and still able to carry out complicated calculations as well as dynamic obstacle detection and avoidance. The key approach towards the project was to explore the various alternatives that are available before committing to a particular component or system. In the case of this project, the Beaglebone Black was used instead of the more popular Raspberry Pi due to the improved specifications and performance. The popular Kinect was also replaced by the Xtion Pro Live as the choice of the vision in the system due to the reduced power requirements of the Xtion Pro Live. The results obtained from the experiments carried out has shown that the Xtion Pro Live mounted robot with the Beaglebone Black single board computer was indeed able traverse and avoid obstacles in the path of motion and that the differential scheme of movement was the most efficient out of the three schemes proposed. Overall, the project has provided a great learning experience in the field of robotics as well as robotic vision. It has provided me with a myriad of information on the different ways that obstacle avoidance can be carried out for any robots with its distinctive set of sensors. It has also allowed me to improve my non technical skills such as planning as well as time and project management.