Robotics and AI

This page provides a very brief introduction to some of the major robotics, artificial intelligence and machine learning-based projects in which I was involved prior to commencing my Ph.D in 2013.

  • For information regarding my current projects in systems and computational biology, please click here
  • For a complete list of my robotics and AI-related research publications, please click here

Gliders: CSIRO’s RoboCup 2D Simulation League team

CSIRO Computational Informatics, 2012-2013

The RoboCup Simulation League abstracts away the hardware limitations of physical robots to allow greater research emphasis on artificial intelligence and team strategy. In the 2D Simulation League, two teams of 11 simulated autonomous software programs (called agents) play soccer in a two-dimensional virtual soccer stadium represented by a central server, called SoccerServer. Each player receives relative and noisy input from its virtual sensors (visual, acoustic and physical) and is able to perform basic commands (like dashing, turning or kicking) in order to influence its environment.

Gliders have completed in the RoboCup 2D Simulation League world finals since 2012. The team has been very successful in this short time, finishing 4th in 2012 (Mexico City), 5th in 2013 (Eindhoven) and 2nd in 2014 (João Pessoa).

During my internship with the CSIRO, I was responsible for the implementation of Gliders’ agent self-localisation system. This system is what allows agents to identify their position and orientation on the virtual field from noisy visual input. My particle filter-based localisation system was deployed for the 2013 competition and is used by the Gliders to this day.

Natural language processing: Shakespearean stylistics

Hunter Medical Research Institute, 2012

Generally speaking, natural language processing (NLP) is a field of computer science, artificial intelligence and linguistics concerned with the interactions between computers and human (natural) languages. Computational stylistics is an area of NLP that aims to find patterns in language that are linked to the processes of writing and reading, and thus to authorial “style” in the wider sense, but that are not demonstrable without computational methods.

While working as a research associate at the Hunter Medical Research Institute in 2012 (under the supervision of Pablo Moscato: inventor of memetic algorithms), I was involved in the development of software capable of accurately identifying the authorship of Shakespearean-era plays. This software is able to extract the distinctive signatures of individual authors by identifying fluctuations in the observed frequencies of common word usage.

Our approach for extracting word usage “signatures” is readily-applicable to the identification of disease “biomarkers” in medical research.

NUbots: The University of Newcastle’s humanoid robotics team

The University of Newcastle, 2011-2012

In the RoboCup Humanoid League, autonomous robots with a human-like body and senses play soccer against each other. Areas of active research in the Humanoid League include: dynamic walking, running and kicking the ball while maintaining balance; visual perception of the ball, other players and the field; self-localization and basic team strategy.

The NUbots are two-time RoboCup world champions and have been active participants since the competition’s inception. I joined the team in 2011 as part of an undergraduate research scholarship with the University of Newcastle’s Interdisciplinary Machine Learning Research Group, and I was fortunate enough to compete at both RoboCup 2011 (Istanbul) and 2012 (Mexico City) with the NUbots.

Over a period of 2 years, the NUbots helped me `learn the ropes’ of robotics. I completed my engineering Honours research project in machine learning and robotics (as of 2012, this was the highest-scoring Honours project ever awarded for my degree by the University of Newcastle) and have since undertaken several research projects in computer vision, locomotion, localisation and robot behaviour.