Lyapunov-Based Control
Overview
The field uses Lyapunov functions—scalar measures that decrease along system trajectories—to prove convergence and design feedback for robots, vehicles, and other nonlinear systems. Control Lyapunov functions encode progress toward goals, while control barrier functions, often combined in quadratic programs or closed-form safety filters, enforce state and actuator constraints. Current work also learns Lyapunov certificates or uses them to guide reinforcement learning, adaptive control, communication timing, and generative flows, with evaluation ranging from simulation to hardware. A central challenge is preserving rigorous guarantees under model error, partial observability, actuator limits, and distribution shifts without making controllers too conservative or computationally expensive.
Latest papers
17 papers1 paper in the last four weeks, against 2 the four weeks before. 0.0% of all new papers.