João Gaspar Cunha, a PhD student in the Doctoral Programme in Electronic and Computer Engineering, is a finalist in the NeuroDesign EXPO & Competition, part of IEEE RO-MAN 2025, which took place in Eindhoven.

This competition aims to find the best projects that exemplify the principles of NeuroDesign innovation to identify, invent and implement developed HRI technologies that can provide practical guidance to quickly take research from the Human-Robot Interaction laboratory to solving real-world problems.

The work that led João Gaspar Cunha to the final of this competition, which is being voted on alongside others, is entitled ‘The Neuroevolution of Collaborative Decision-Making in Robotic Assistants’ and was supervised by Estela Bicho (EEUM), Raymond Cujpers and Wolfram Erlhagen.

The work focuses on how robots can learn to collaborate naturally with humans, exploring how the principles of Neuroevolution and Dynamical Field Theory can give rise to adaptive and interpretable control in robotic assistants. By developing neural architectures that govern perception, memory, and decision-making, our system allows robots to learn when to help and when to act independently in shared tasks. In a human-robot packaging scenario, the evolved controllers exhibit emergent collaboration, complementarity, and self-organisation without manual adjustment. This work proposes a new generation of evolved neuroadaptive robotic partners capable of fluid, human-like collaboration shaped by the principles of the brain.

Project video here

See all the finalist projects and vote at https://www.neurodesign-hri.ws/2025#final_round_2025