SCS Undergraduate Thesis Topics
|Sonia Chernova||Manuela Veloso||Adaptive Motion for Quadruped Robots|
Robotic soccer is a complex task that requires multiple autonomous agents to collaborate in an adversarial environment to achieve specific objectives. The Sony quadruped robots are fully autonomous with onboard computation, visual perception, and sophisticated head, leg, and tail motions. The robotic soccer task requires a large suite of different motions, including straight walk, turns, kicks, and handling of obstacles (walls and other robots). The goal of this research is to implement an algorithm to evaluate the effects of different kick motions and to select an optimal strategy in a play situation. Additionally, I will look at learning algorithms that would enable the Sony AIBO robot to adapt its gait over time and optimize the velocity and stability of its motion on a given surface.