Research project

FAST-PI: Flexible Autonomy in Human-Swarm Teams Using Real-Time Physiological Information

Project overview

This project will investigate new approaches to operate and evaluate human-swarm systems with flexible autonomy, thus establishing new metrics and testing methodologies for such systems. We will do so by analysing the physiological activity of human operators and adjusting swarm autonomy accordingly. Operators’ neurophysiological responses (e.g., ErrPs) will be analysed to measure situational trust as they identify casualties (in a disaster response scenario) with a UAV swarm. Error-related potentials are generated in the human brain when they perceive undesirable behaviour and will assist in detecting precursors of trust. In response, the swarm will adapt its autonomy and the presentation of explanations. Higher trust leads to higher levels of swarm autonomy and lower trust will allow operators to intervene and manually pilot UAVs.

Staff

Lead researchers

Dr Mohammad Soorati Dr.Eng., MSc, BSc

Lecturer

Research interests

  • Multi-robot Systems
  • Human-swarm Interaction
  • Explainable AI 
Connect with Mohammad

Research outputs