Available Projects (2024)

The degenerate Kalman filter – Read more

Gaussian belief space planning using probabilistic graphical models – Read more
Co-supervisor: Corné van Daalen

"Diplomacy" using reinforcement learning – Read more
Co-supervisor: Herman Kamper

Using reinforcement learning to improve human learning in arcade games – Read more
Main supervisor: Herman Kamper

Current Students

Andrew Murdoch (PhD)
Co-supervisor: Willem Jordaan
A learning-based approach for competitive multi-vehicle autonomous racing

Francois Bredell (PhD)
Co-supervisor: Herman Engelbrecht
Learning to play Hanabi using multi-agent reinforcement learning

Devin Jefferies (MEng)
Co-supervisor: Benjamin Evans
Autonomous F1tenth racing on an unknown track

Emile Visser (MEng)
Co-supervisor: Corné van Daalen
Improving the accuracy of Bayesian optimisation using an adaptive approach

Stephano Buys (MEng)
Co-supervisor: Japie Engelbrecht
Cooperative planning for autonomous search and rescue

Past Students

2023

Andrew Murdoch (MEng)
Co-supervisor: Willem Jordaan
Partial End-to-End Reinforcement Learning for Autonomous Racing

Ulrich Louw (MEng)
Main supervisor: Willem Jordaan
Autonomous Diagnosis of Satellite Sensor Anomalies to Ensure Fault Tolerant Control – View thesis

Welri Botes (MEng)
Main supervisor: Japie Engelbrecht
Grid-Based Coverage Path Planning for Multiple UAVs in Search and Rescue Applications – View thesis

2021

Cobus Louw (MEng)
Main supervisor: Herman Engelbrecht
Solving Sparse-reward Problems in Partially Observable 3D Environments using Distributed Reinforcement Learning – View thesis