Mohammad Divband Soorati

Research interests

Multi-robot Systems

Human-swarm Interaction

Explainable AI

Accepting applications from PhD students.

About

Dr Soorati is an Assistant Professor of AI at the University of Southampton. He is interested in human-swarm interaction, swarm robotics, trust in human-swarm teaming, aerial swarms, and evolutionary/bio-robotics. He is sitting on the board of Center for Machine Intelligence and Center for Robotics at the University of Southampton. His research has been covered by several international news outlets (including EuroNews, France24 and Deutsche Welle) and magazines (such as IEEE Spectrum, Robohub and Future Maisons). He is a co-investigator of an EPSRC project on 5G-connnected UAVs and project lead on eXplainable Human-swarm Systems.

He is an associate editor of ICRA, an academic editor of the PLOS ONE Journal, sponsor chair of ANTS, a member of the program committee of many high-impact conferences/journals in AI & robotics such as IJCAI, AAAI, IROS, AAMAS, Swarm Intelligence Journal, etc.

Research

Research groups

Member of:

Agents, Interaction and Complexity

Research interests

Multi-robot Systems
Human-swarm Interaction
Explainable AI
Trustworthy AI
Machine Learning in Multi-robot Systems

Current research

Human-swarm Interaction: This domain addresses the interaction between humans and multi-agent systems. The goal is to develop effective tools and methods for human operators to interpret and manage multi-agent system actions, ensuring both operational efficiency and adherence to ethical and safety standards.

Swarm Robotics: The research in this area focuses on the design and development of hardware and controllers for robot swarms, enabling them to autonomously execute complex tasks.

Multi-agent Systems: Mohammad's research revolves around studying systems with multiple agents. Machine learning techniques are employed to analyze and predict agent behaviors and intentions.

Distributed Machine Learning: A significant part of the research is dedicated to distributed machine learning, which enables the analysis of data spread across multiple devices or nodes. This approach is particularly valuable for designing controllers in real-time systems and for understanding the intricacies of complex natural organisms. By dissecting the inherent mechanisms within these organisms, we can develop more efficient and intelligent multi-agent systems.