Postdoctoral Research Associate
Postdoctoral Research Associate in Safe Autonomy of Multi-Robot Systems
The position is funded by the Horizon Europe project “REXASI-PRO: Reliable & Explainable Swarm Intelligence for People with Reduced Mobility”. The project aims to develop a framework for safe and trustworthy social navigation of multi-robot systems, with application (among others) to swarms of autonomous wheelchairs and drones.
Within this framework, you will work under the supervision of Dr Nicola Paoletti to develop model-based probabilistic verification methods for such multi-agent systems, to enable both design-time and run-time safety assurance.
Key challenges involve dealing with systems with machine-learning components in the loop (used for perception and end-to-end control), effectively scaling up the methods to support multiple agents, and accounting for different uncertainty and stochasticity sources, due to e.g., partially observable and evolving environments, unreliable learning models, or inaccurate sensing and actuation. To tackle these challenges, you will investigate novel methods for compositional verification and predictive monitoring of learning-enabled systems, as well as for uncertainty quantification, with an emphasis on conformal prediction.
Dr Paoletti has extensive experience in formal analysis of cyber-physical systems. His recent research work includes reach-avoid verification of Bayesian neural networks and conformal prediction for predictive monitoring of cyber-physical systems, topics that are highly relevant for this position. You will also collaborate with the project’s academic and industrial partners (from Italy, Germany, Spain, Switzerland, and Belgium), with a combined expertise spanning from robotics to AI ethics.
This post will be offered on an a fixed-term contract until 30th September 2025
This is a full-time post - 100% full time equivalent