Adaptive Control of Nonlinear Systems

Adaptive control of nonlinear systems is concerned with designing feedback controllers that can adjust their own parameters in real time as a system's dynamics change or remain partially unknown. Unlike classical control methods that assume a fixed, well-characterized model, adaptive approaches allow robots, aerial vehicles, and other complex machines to maintain stable, precise behavior in the presence of uncertainties, unmodeled disturbances, and varying operating conditions. Techniques such as sliding mode control, disturbance observers, and neural network-based approximators have emerged as practical tools for achieving this robustness, with finite-time stability guarantees becoming an increasingly important benchmark for safety-critical applications like quadrotor flight and robotic manipulation. Open questions include how to make these controllers computationally light enough for real hardware constraints, and how to provide formal guarantees when learning-based components are integrated into the control loop.

Works

84,336

Total citations

1,476,897

Keywords

Adaptive ControlSliding Mode ControlDisturbance ObserverNonlinear SystemsFinite-Time StabilityQuadrotor