Final Year Project

Project Title

Quadruped Locomotion in Rough Terrain and Confined Environments

Description

Legged robots are used for inspection, exploration, and search and rescue missions in tough environments such as caves. In our capstone project, we sought to make a hobby-grade legged robot kit capable of traversing rough terrain and navigating through confined spaces. We upgraded the hardware and developed a simulation package for the robot. We used classical control techniques, learning-based approaches, and evolutionary algorithms for locomotion in rough terrain, along with computer vision techniques to perceive obstacles and decide on a posture that was used to constrain the locomotion algorithms to ensure the robot moved with a height and width that did not collide with the environment. Thanks to our simulation package, we were able to rapidly test our algorithms.