Aidan Scannell

Email: scannell.aidan@gmail.com

PhD Project Title: Investigating the Application of Bayesian Model-Based Reinforcement Learning on Physical Robotic Systems

Research Group: Aerial Robotics

Website: Aidan Scannell

Biography

I grew up in North Yorkshire and graduated from the University of Bristol in 2016 with an MEng in Mechanical Engineering. In 2017 I moved out to Whistler, BC, and spent the year snowboarding in the mountains; it was a great year! My masters thesis was focused on extending an agent based programming language (AgentSpeak(L)) to model and reason with uncertainty, specifically, extending agents’ with probabilistic and possibilistic beliefs. I am now conducting my PhD under the supervision of Professor Arthur Richards and Dr Carl Henrik Ek, focusing on data efficient learning for robotic systems (quadcopters). I am particularly interested in uncertainty quantification in model-based reinforcement learning, and as a result a lot of my work focuses on Bayesian non-parametric methods, specifically Gaussian processes. I am also very interested in methods for encoding expert domain knowledge into such models. Outside of my work I spend a lot of time climbing up walls and snowboarding down mountains.

Website: https://www.aidanscannell.com

Research Interests

- Reinforcement learning

- Gaussian processes

- Bayesian inference

- Probabilistic modelling

- Robust control