Behçet Açıkmeşe; Convexification-Based Real-Time Optimization for High Performance Autonomous Control

May 31, 2016, 4pm
GUG 204
Behçet Açıkmeşe, Department of Aeronautics & Astronautics, University of Washington

Abstract: Many future engineering applications will require dramatic increases in our existing Autonomous Control capabilities. These include robotic sample return missions to planets, comets, and asteroids, formation flying spacecraft applications, applications utilizing swarms of autonomous agents, unmanned aerial, ground, and underwater vehicles, and autonomous commercial robotic applications. A key control challenge for many autonomous systems is to achieve the performance goals safely with minimal resource use in the presence of mission constraints and uncertainties. In principle these problems can be formulated and solved as optimization problems. The challenge is solving them reliably onboard the autonomous system in real time.

Our research has provided new analytical results that enabled the formulation of many autonomous control problems in a convex optimization framework, i.e., convexification of the control problem. The main mathematical theory used in achieving convexification is the duality theory of optimization. Duality theory manifests itself as Pontryagin’s Maximum Principle in infinite dimensional optimization problems and as KKT conditions in finite dimensional parameter optimization problems. Both theories were instrumental in our developments. Our analytical framework also allowed the computation of the precise bounds of performance for a control system in term of constrained controllability/reachability sets. This proved to be an important step in rigorous V&V of the resulting control decision making algorithms.

This presentation introduces several real-world aerospace engineering applications, where this approach either produced dramatically improved performance over the heritage technology or enabled a fundamentally new technology. A particularly important application is the fuel optimal control for planetary soft landing, whose complete solution has been an open problem since the Apollo Moon landings of 1960s. We developed a novel “lossless convexification” method, which enables the next generation planetary missions, such as Mars robotic sample return and manned missions. Another application is in Markov chain synthesis with “safety” constraints, which enabled the development of new decentralized coordination and control methods for spacecraft swarms.

Biographical Sketch: Behçet Açıkmeşe is a faculty in Department of Aeronautics and Astronautics at University of Washington, Seattle. He received his Ph.D. in Aerospace Engineering from Purdue University. Previously, he was a senior technologist at JPL and a lecturer at Caltech. At JPL, Dr. Açıkmeşe developed control algorithms for planetary landing, spacecraft formation flying, and asteroid and comet sample return missions. He was the developer of the “flyaway” control algorithms in Mars Science Laboratory (MSL) mission, which successfully landed on Mars in August 2012, and the reaction control system algorithms for NASA SMAP mission, which was launched in 2015. Dr. Açıkmeşe is a recipient of NSF CAREER Award, several NASA Group Achievement awards for his contributions to MSL and SMAP missions, and to Mars precision landing and formation flying technology development. He is an Associate Fellow of AIAA, a senior member of IEEE, and an associate editor of IEEE Control System Magazine and AIAA JGCD.