Towards Bio-inspired Estimation and Control for Autonomous Systems: Lessons from Collective Fish Behavior
Collective animal behavior provides numerous examples of robust coordination and decision-making in complex environments. Fish schools, in particular, perform navigation and aggregation tasks by integrating noisy multisensory cues and propagating information across the group. This makes them a natural model system for studying decentralized estimation and control problems in multi-agent systems. This talk will present recent theoretical advances in modeling and analysis of fish behavior through the lens of dynamical systems and control theory. I will introduce a data-driven framework, anchored in stochastic differential equations, that describes fish dynamics while incorporating hydrodynamic and visual feedback. Building on these models, I will show how fish-inspired principles translate into practical tools for: (i) autonomous robot navigation in unknown environments via zero-order optimization and (ii) social decision-making modeling with analytic predictions of collective outcomes. I will conclude by outlining emerging directions in bio-inspired algorithms for inference and control.
Bio: Daniel Burbano is an assistant professor of electrical and computer engineering at Rutgers University, where he leads the Swarm Intelligence Laboratory. His research interests encompass the analysis and control of complex systems and networks, with a focus on data-driven modeling, collective decision-making in animal collectives, and bio-inspired autonomy. He received the B.Sc. in electronic engineering and the M.Sc. in industrial automation from the National University of Colombia (2010, 2012), and the Ph.D. in control engineering from the University of Naples Federico II, Italy (2015). He was a postdoctoral researcher at Northwestern University and a Faculty Fellow at New York University. Burbano is the recipient of the NSF CAREER Award. He serves as a Senior Associate Editor for IEEE Transactions on Circuits and Systems II (TCAS II) and as an Editor for PLoS Complex Systems.