I am a Lecturer in Robotics at Monash University in Melbourne, Australia. Prior to this, I was a Research Associate at the Institute of Perception, Action and Behaviour at the University of Edinburgh, which I joined after returning to academia in 2018. Before that, I led a team of 20 staff and students working in computer vision, machine learning and field robotics at the Mobile Intelligent Autonomous Systems group at the Council for Scientific and Industrial Research (CSIR), South Africa, which I joined in 2009. I have a PhD in statistical signal processing from the University of Cambridge (2012-2016), a Masters of Science in electronic engineering from Stellenbosch University and a Bachelors in electronic engineering from the University of Pretoria.
Recruiting: I am actively looking for PhD students interested in probabilistic machine learning and inference for robotics, vision-based robot control, computer vision, robot learning and field robotics. Drop me an email if you’re interested in applying (see https://www.monash.edu/graduate-research/future-students/apply for application details).
This website lists projects I’ve worked on in a variety of areas covering robotics, computer vision and machine learning, particularly the probabilistic kind. I have also been involved in a wide range of contract R&D projects covering areas ranging from agriculture to mining, but sadly I can’t show these off here. At present my work focuses on building interpretable and verifiable learning architectures for safe and robust robot autonomy, and lightly supervised behaviour learning with hybrid dynamical systems.
Recently, we have been exploring the use of time as a supervisory signal for learning from demonstration. As an example use case, we considered ultrasound scanning, where a technician is required to search for a scanning position and contact force that produces an optimal image. We propose a probabilistic temporal ranking (PTR) model that allows […]
Standard architectures for neural networks have numerous problems with interpretability, flexibility and generalisation. I believe that this is in large part due to a lack of stronger inductive biases in models and architectures, and have recently been pushing (see my job talk at Monash) to include stronger biases in deep learning models. Switching controller front-ends […]
It’s well established that machine learning has a problem with bias. Our datasets reflect the inequalities and prejudices of our daily lives, and the models we train and deploy exacerbate these even further. We even notice this in robotics (despite being generally removed from people), where localisation and mapping systems perfected in green European settings […]
Hybrid system identification can be particularly challenging, particularly in the context of visuomotor control. We introduce switching density networks (SDNs), which can be used to identify switching control systems in an end-to-end learning fashion from demonstration data. We show that SDNs, when paired with a general purpose family of proportional-integral-derivative control laws, can identify the […]
Click here to see more.
I regularly supervise students and staff working on a range of projects in robotics, computer vision and machine learning. At the moment, these include:
- Miguel Jaques – Physics as inverse graphics (co-supervising with Tim Hospedales at the University of Edinburgh, PhD)
- Arturas Straizys – Precision cutting of soft deformable materials (co-supervising with Subramanian Ramamoorthy at the University of Edinburgh, PhD)
- Zimkhitha Sijovu – Probabilistic state estimation and calibration for robot manipulators (co-supervising with Dr Corné van Daalen at Stellenbosch University, MEng)