Nonlinear chance-constrained optimal control under dynamic uncertainty for safety-critical systems

Sponsor: Purdue University

This project develops a control theoretical framework to achieve robust operations of dynamic systems with formal guarantees of their performance under uncertainty. Such performance guarantees are particularly crucial for safety-critical systems, e.g., aerospace systems and autonomous ground vehicles. Major challenges in providing performance guarantees for real-world systems include the handling of nonlinear dynamical systems under uncertainty (i.e., nonlinear stochastic systems) and the existence of structured uncertainty in the dynamical environments (e.g., non-Gaussian uncertainty). This project aims to advance the theory of chance-constrained optimal control to address these challenges by integrating techniques in the fields of uncertainty quantification and optimal control.