Daniel Hernandez

Daniel Hernandez

Multi-Agent Reinforcement Learning for Game AI and Robotic Control

The Internet of Things and connected devices are creating opportunities for increased automation. We are progressing to a point where everything is connected (from the smallest sensor to self-driving cars and aerial drones) but the way we interact with each of these is currently limited to one-one interactions.

Daniel’s project researches and develops multi-agent reinforcement learning algorithms for controlling swarms of agents, enabling an individual to control vast numbers of connected devices simultaneously.

These algorithms will initially be tested in Starcraft to enable rapid iteration. Using a robotic platform provided by Accelerated Dynamics, a London based robotic start-up, these algorithms will be tested in real world scenarios with swarms of aerial drones.

 

Daniel received an MEng in Computing: Games, Vision & Interaction from Imperial College London. Wanting to combine the power of AI and the creativity of videogames, Daniel began a PhD journey to explore the misty lands of Multi Agent Reinforcement Learning (MARL).

Home institution: York

Supervisors: Dr Sam Devlin, Professor Peter Cowling

Ready to apply?

Once you have identified your potential supervisor, we would encourage you to contact them to discuss your research proposal.

Learn More