Amanda Prorok is Professor of Collective Intelligence and Robotics in the Department of Computer Science and Technology at the University of Cambridge, and a Fellow of Pembroke College.

In her work, she pioneered differentiable communications methods for multi-agent systems, with applications to multi-robot perception and control. Amanda has given invited keynotes at TEDx and IEEE ICRA, and has been honored by numerous research awards, including a prestigious ERC Starting Grant.

Amanda is an IEEE Senior Member, serves as Associate Editor for Autonomous Robots (AURO) and was the Chair of the 2021 IEEE International Symposium on Multi-Robot and Multi-Agent Systems. Her PhD thesis was awarded the Asea Brown Boveri (ABB) prize for the best thesis at EPFL in Computer Science. 

Ajay’s research is that of a full-stack roboticist – with a focus on robust, optimal, and agile control + planning for various robots and robotic teams. Current focus is on scalable and learnt multirobot coordination.

Sally works with the team coordinating the lab space, events, resources and grant administration.

Zhan works on understanding the relationship between the multi-agent system and the environment and exploits the latter to improve the system performance. For research, he focuses on techniques in the field of machine learning, particularly reinforcement learning and graph neural networks. 

Jasmine is interested in research at the intersection of mathematics, machine learning theory, and robotics that can enhance the way agents learn and interact. She focuses on parameterising uncertainty and strengthening resilience.

Michael’s research is in multi-agent systems and swarm robotics, focusing on scalability and provable safety guarantees. He is particularly interested in how interactions between an increasing number of independent systems can scalably improve decision making or make it safer. Michael is also interested in alignment.

Jan’s research is about transferring Multi-Agent control policies trained in simulation to the real world (sim-to-real transfer), using Multi-Agent Reinforcement Learning and Graph Neural Networks. He is also interested in interpretability, resilience and robustness of such control policies, particularly in the context of real-world systems.

Sridhar works on applied robotics and machine learning. He has an interest in moving cutting-edge research to real-world applications. Across experiences in startups, industry, and academia, he has worked on adapting and deploying research in computer vision, imitation learning, and reinforcement learning to specific use cases.

Mateusz’s research focuses on Multi-Agent Reinforcement Learning, particularly on representing the intentions of agents in environments without communication. He is also interested in exploring how heterogeneity in agent systems shapes coordination and learning dynamics.

Rishabh’s work aims to make GNNs safer, with an emphasis on multi-agent systems. This includes making (certifiably) robust models and exploring explainable and interpretable models.

Keisuke has broad interests in controlling multiple moving agents. In particular, his research focuses on path planning, combinatorial search, learning and distributed systems.  He is a JSPS Overseas Research Fellow in Japan.

Read more

Eduardo works on the development of multi-robot decision-making systems driven by high-level task instructions. Specifically, he is interested in distributed algorithms that allow task decomposition, collective adaptation and seamless integration of the multi-robot system into the environment.

Antonio’s research focuses on developing learning-based control and coordination strategies for large-scale, heterogeneous multi-agent systems. He is particularly interested in using compositional methods to integrate control theory with machine learning, ensuring provable guarantees on system-wide stability, robustness, and scalability.

Maksymilian’s goal is to enable teams of robots to reason collaboratively in a scalable and decentralised manner. He is also interested in topics such as zero-shot coordination, world models and open-endedness.

Jennifer works on understanding wireless data communications mechanisms and their applications to multi robot control algorithms, studying strategies and co-optimizations that allow predictable transfer to the real world. Her work may be applied to automotive inter-networking and drone swarm control.

Peter is pioneering approaches that move beyond monolithic AI models, focusing instead on specialized, collaborative robot teams that can adapt dynamically and scale efficiently. His work bridges theoretical advances in modular AI with real-world experiments involving drone swarms and multi-robot systems.