Andreas Themelis

Papers

Total Citations

H-Index

About

Andreas Themelis is a leading figure in optimization-based control, with his research primarily centered on real-time numerical methods for nonlinear model predictive control (NMPC) and optimal control. His most significant contribution is the development of PANOC (proximal averaged Newton-type method for optimal control), a groundbreaking algorithm that enables fast, embedded solutions to complex control problems. Themelis’s 2018 paper on using PANOC for obstacle avoidance, cited 98 times, demonstrates his ability to bridge theory and practice by introducing a novel modeling framework that handles generic, nonconvex obstacles—such as polytopes and ellipsoids—in real time. This work has profound implications for autonomous systems, including drones and mobile robots, where rapid decision-making is critical. Beyond this, Themelis has advanced the fields of distributed optimization and stochastic control, with his research consistently achieving high impact through both theoretical rigor and practical applicability. His contributions have been recognized with prestigious awards, including the IEEE CSS George S. Axelby Outstanding Paper Award, cementing his reputation as a pioneer in making advanced control algorithms computationally viable for real-world deployment.