Robust Controllable Set

Robust controllable sets aim to design systems that reliably function despite uncertainties in their environment or internal parameters. Current research focuses on developing efficient algorithms, often leveraging constrained zonotopes and least-squares methods, to compute these sets, particularly for high-dimensional systems and under various types of uncertainty (e.g., adversarial attacks, dataset shifts). This work is crucial for building reliable AI systems and control strategies in safety-critical applications, such as autonomous vehicles and spacecraft navigation, where robustness to unforeseen events is paramount.