Our research is aimed at designing and controlling real-time, distributed systems consisting of autonomous units that can perceive, decide, collaborate, and act in the physical world. Our work addresses system engineering both at the hardware and software levels, model-based and machine-learning-based analysis and synthesis methods with particular emphasis on distributed, scalable control algorithms. Swarm-intelligent systems, and in particular swarm-robotic ones, exploit self-organization as the key coordination mechanism, are statically and dynamically scalable, and are characterized by an often large number of units with limited individual complexity in respect to the task they are supposed to accomplish.
This project focuses on a specific class of problems: distributed sensing. In particular, it aims at: first, generalizing problem-specific analysis and synthesis solutions to engineering methodologies; second, understanding the theoretical limits of a swarm-robotic approach for distributed sensing problems; and third, comparing the swarm-robotic solution with other more standard engineering methods in terms of efficiency and robustness as well as system cost. These issues are addressed with the following three strongly interconnected projects, each of them involving 1 PhD student: (A) Distributed Boundary Coverage of Regular Structures using Miniature Swarm-Robotic Systems; (B) Distributed Localization of Acoustic Targets Using Networked Swarm-Robotic Systems; (C) Distributed Resource Allocation and Attention in Networked Swarm-Robotic Systems. The three projects maximize cross-fertilization by often sharing experimental, computational, and mathematical tools while maintaining a strong individual signature and complementary goals. The efforts in the development of project-independent tools and methods will be further strengthened in this project with the help of a part-time postdoctoral fellow.