Automatic department seminar "Distributed multi-robot control: Physics, geometry and learning"

Multi-robot systems are emerging as a promising solution for tackling complex tasks that exceed the capabilities of a single robot. Their inherent parallelism, resistance to individual failure and ability to operate in large-scale environments make them particularly attractive for applications such as search and rescue, environmental monitoring, animal husbandry and agriculture, or warehouse automation.

However, coordinating and controlling multiple robots operating in a distributed infrastructure poses significant challenges. Specifically, this talk addresses three key aspects of distributed multi-robot systems: efficient coordination in highly non-linear and volatile environments, fast and accurate reconstruction of collective information, and scalability of control policies as a function of the number of robots.

The first aspect is the need to ensure that the control policy is scalable to the number of robots.

To achieve this, the work presented in this talk exploits three main tools: the physical properties of networked systems, geometric control techniques and distributed optimization and machine-learning methods.

Eduardo is a postdoctoral researcher at ProrokLab, University of Cambridge.

He obtained his PhD in the Robotics, Computer Vision and Artificial Intelligence group at the University of Zaragoza, under the supervision of Eduardo Montijano and Carlos Sagüés.

He was a visiting researcher at UCSD's Existential Robotics Laboratory in 2022 and 2024.

His research focuses on topics related to multi-robot decision-making, networked systems, control and learning (and, occasionally, power electronics).

Eduardo holds a Fulbright scholarship and a DAAD AInet fellowship.