Research

Multi-Agent Systems

We study decision-making, coordination, and control in networked and large-scale multi-agent systems. A central focus is the development of scalable and distributed methods that allow multiple autonomous agents to cooperate, compete, and accomplish shared or individual objectives under limited information and uncertainty. Our research draws on control theory, game theory, optimization, and learning to address problems such as distributed coordination, cooperative and competitive decision-making, formation and swarm behavior, and large-scale multi-agent optimization. Applications include multi-robot and networked autonomous systems involving aerial and ground robotic platforms.

Optimal and Learning-Based Control

We develop control and decision-making methods for complex dynamical systems by combining model-based optimization with data-driven and learning-based approaches. Our research investigates how optimal control, adaptive control, reinforcement learning, and related computational methods can be used to improve performance and adaptability when accurate models or complete environmental information are not available. An important aspect of this work is understanding how learning can complement established control and optimization methods rather than replacing them. We are interested in methods that balance performance, computational requirements, model knowledge, and adaptability for autonomous and robotic systems.

Safe and Resilient Autonomous Systems

We investigate methods for enabling autonomous systems to operate reliably in uncertain, dynamic, and unstructured environments. Our research considers safety and resilience in the presence of model uncertainty, external disturbances, changing operating conditions, limited control authority, and other system or environmental constraints. We combine control-theoretic safety and stability methods with optimization and learning to develop autonomous decision-making and control strategies that can adapt to changing conditions while respecting operational requirements. Current applications include aerial and ground robotic systems and other safety-critical autonomous platforms.