Date of Award

Spring 2024

Document Type

Thesis

Publisher

Santa Clara : Santa Clara University, 2024

Department

Electrical and Computer Engineering

First Advisor

Maria Kyrarini

Second Advisor

Ahmed Amer

Abstract

RoboCup is an annual international robotics competition, established in 1997, aimed at promoting research and education in artificial intelligence and robotics through a variety of competitive leagues. One of the key leagues is the Standard Platform League (SPL), where all teams use identical robots—commonly the NAO robot by SoftBank Robotics. This ensures a focus on software innovation and strategic development. Typically held in July, RoboCup attracts hundreds of teams from around the world, with the SPL matches featuring teams of up to five robots per side. The competition fosters collaboration and advancements in Artificial Intelligence (AI), as teams develop sophisticated algorithms for autonomous decision-making and teamwork shown in the figure 1.1.

Also, robots today are getting more advanced and required to work on more sophisticated projects than ever before. Due to this increasing complexity of projects that need to be done by robots, robots that can initiate complex movements are only going to increase as time goes on. However, as of right now, creating such robots is hard to both manually program and test, requiring many weeks, if not months, to train and test a robot to do complex motions like walking over rough terrain, or dancing. However, with the rise of artificial intelligence and machine learning, a new way of training these robots has appeared, and our project is also designed to take advantage of this new technology.

The initial motivation of this project was to get Santa Clara University to compete in the RoboCup Standard Platform League. We wish our development of the training technologies and Machine Learning (ML) applications on the NAO robot together with more robots and funding in the future could earn SCU a spot in the RoboCup competition.

Download

Share

COinS