Exploiting Asymmetry for the Control of Multi-Agent Systems
Abstract
Recent research has identified scenarios in which on otherwise unstable symmetric state can be stabilized in a network if the given symmetry is broken in the dynamical equations of the system. This effect, referred to as converse symmetry breaking, has implications for stability in systems with tunable parameters. In particular, it shows that the optimal parameter assignment is not necessarily limited to a low dimensional subspace satisfying the symmetry of the state of interest. This creates new challenges and opportunities for real-time control of multi-agent systems, which we propose to explore using a combination of mathematical and computational approaches. In this STIR project, we plan to explore converse symmetry breaking to develop an improved online feedback control approach for swarming robots. Swarming of unmanned aerial, ground, and underwater vehicles has gained significant attention as alternative that offers cost and performance advantages for tasks such as surveillance and target location. The collective behavior in such engineered systems is intended to emerge from the interactions among the agents and between the agents and the environment. An implicit assumption usually made is that identical agents facilitate coordination. This project builds on preliminary evidence that specific forms of inter-individual differences could in fact facilitate coordination in swarming robotics-a manifestation of converse symmetry breaking in these systems. In this project, we will establish an approach for increased stability of swarming robots to facilitate applications to problems of relevance to the Army, such as target localization, surveillance, and transport with limited communication capabilities. The proposed research is inherently interdisciplinary and will benefit from Northwestern University s tradition in interdisciplinary complex systems research. The project will involve a graduate student and postdoctoral researcher, and it will thus contribute to the formation of the next generation of interdisciplinary scientists.
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- Apr 19, 2023
- Source ID
- W911NF2310102
Entities
People
- Adilson E. Motter
Organizations
- Army Contracting Command
- Northwestern University
- United States Army