Napat Sahapat

Research Student -  PhD

Research Program: International Centre for Neuromorphic Systems

Thesis Title

Predator Prey Drone Navigation with Reinforcement Learning

Research Project

This project aims to use reinforcement learning to replicate the behaviour of predators and preys found in ecology, i.e. pursuit and evasion behaviours. Drones are used for control of the agents. Using cooperative and competitive multi-agent reinforcement learning, as well as spiking neural networks for more energy-efficient methods, this study aims not only to replicate these behaviours in simulation, but also to create realistic and effective drones that can chase other drones in the sky, as well as work cooperatively with other similar drones.

The project is still in its early stages of research, however; so far, we are only working on a 2D simulation of the predator prey drones.

Qualifications

• Master of Data Science (Western Sydney University)
• Bachelor of Marine Science (Macquarie University)

Contact