On May 30t^h, 2025 at 11.00 am the seminar titled "Three Lessons on Heterogeneity in Robot Swarms" will take place at DEIB Beta Room (Building 24).

Heterogeneity in robot swarms presents both significant challenges and unique opportunities for multi-robot coordination. This seminar explores three key perspectives on heterogeneity. First, we examine heterogeneity as a challenge in multi-robot task allocation. When robots possess different skill sets but tasks require specific combinations of these skills, how can we efficiently form the right teams at the right time? We present our approach to simultaneous task scheduling and coalition formation, demonstrating how distributed optimization can dynamically assemble robot teams with complementary capabilities to tackle complex tasks requiring multiple skills. Second, we investigate heterogeneity as an emergent property in collective transport scenarios. When robots must coordinate to move large objects, uniform behavior often leads to inefficiency and deadlocks. We show how our approach using swarm state prediction enables robots to autonomously differentiate their roles, leading to spontaneous yet coordinated behavioral specialization that significantly enhances transport efficiency. Third, we confront heterogeneity as an unavoidable reality in long-term missions. As robots operate over extended periods, their sensors degrade at different rates, creating performance disparities across the swarm. We present our innovative methods for collective perception that maintain swarm-level sensing capabilities even when individual robots experience severe sensor degradation. Finally, as a bonus perspective, we reframe heterogeneity as an opportunity through the concept of extended stigmergy in swarm construction. We demonstrate how treating structures under construction as active matter creates a powerful coordination mechanism where the emerging structure itself guides seemingly simple construction robots through complex building processes.

Carlo Pinciroli is an associate professor at Worcester Polytechnic Institute, where he leads NEST Lab. His research interests include multi-robot systems, swarm intelligence, and software engineering. Dr. Pinciroli published more than 70 peer-reviewed articles and 2 book chapters, and edited 2 books. He is associate editor for the Swarm Intelligence journal and for the IROS and ICRA robotics conferences. Dr. Pinciroli was the recipient of several awards for research and teaching, including a Best Paper Award at AAMAS 2022, WPI's Romeo L. Moruzzi Young Faculty Award for Innovation in Undergraduate Education (2020), and an Amazon Research Award (2019).