Abstract
In this talk, I will explore the role of behavioural similarity assessment in swarm robotics and its relevance to imitation learning, and review existing approaches proposed by the community. However, these approaches typically address narrow behavioural contexts, making it difficult to establish a general methodology or unifying framework for swarm roboticists. Consequently, I will present a range of behavioural similarity assessment methods, including feature-based and generative adversarial approaches, and discuss how they can support more systematic imitation learning in swarms.
About the speaker
I am a Ph.D. researcher in the Robot Swarm Learning group at the University of Konstanz where I’m affiliated with the Centre for the Advanced Study of Collective Behaviour. My research focuses on imitation learning for robot swarms, drawing inspiration from natural swarm intelligence. Previously, I worked as a Robotics Researcher at the Technology Innovation Institute in Abu Dhabi. I hold a Master’s degree in Artificial Intelligence from Vrije Universiteit Amsterdam.