Engineering Scalable Multi-Robot Systems for Reliable Operation Under Uncertainty and Resource Constraints
This talk presents a constraint-driven engineering perspective on achieving reliable operation in scalable multi-robot systems under uncertainty and resource constraints.
Overview
Robotics has transformed many aspects of modern life, from manufacturing to logistics, yet its impact on large-scale challenges - such as wildfires, locust invasions, and crop epidemics - remains limited. Such scenarios are time-critical and unfold in unstructured environments: they span vast regions, evolve rapidly, and require coordinated action at scale. Addressing them calls for multi-robot systems that can operate concurrently across large areas. However, designing such systems is difficult: no robot has complete information, each must act on local and uncertain observations, and both environmental conditions and hardware performance change over time. In addition, bandwidth, energy, and cost cannot scale linearly with fleet size, and as robot numbers grow they can interfere with one another, making performance less reliable and harder to anticipate. This talk presents a constraint-driven engineering perspective on achieving reliable operation in scalable multi-robot systems under uncertainty and resource constraints. It illustrates this perspective through three engineered systems: collective decision-making in dynamic environments, spatial effort coordination under limited communication, and capability-aware coordination in heterogeneous UAV fleets. By designing with constraints rather than assuming them away, this perspective helps move multi-robot systems from demonstrations to deployable autonomy.
Presenters
Dr. Mohamed Salaheddine Talamali, Research Associate, Energy Aware Swarm Programming, School of Electrical and Electronic Engineering (EEE), University of Sheffield (TUOS)
Brief Biography
Dr. Mohamed Salaheddine Talamali is a researcher in swarm robotics at the University of Sheffield, where he develops decentralized approaches for long-term autonomy in robot swarms, with an emphasis on energy-aware coordination and adaptive learning under uncertainty and resource constraints. Previously, he held a teaching-focused Assistant Professorship in Automation, Control, and Robotics and received the Inspirational Teaching Award twice. He was also a Research Fellow in the Department of Computer Science at University College London, where he applied swarm robotics principles to the design of spatial sound modulators, enabling applications such as acoustic levitation and mid-air haptics. Dr Talamali earned his Ph.D. from the University of Sheffield, focusing on scalable algorithms for large robot swarms operating under limited sensing, communication, and computation.