With this brief, the authors present algorithms for model-free stabilization of unstable dynamic systems. An extremum-seeking algorithm assigns the role of a cost function to the dynamic system's control Lyapunov function (clf) aiming at its minimization. The minimization of the clf drives the clf to zero and achieves asymptotic stabilization. This approach does not rely on, or require knowledge of, the system model. Instead, it employs periodic perturbation signals, along with the clf. The same effect is achieved as by using clf-based feedback laws that profit from modeling knowledge, but in a time-average sense. Rather than use integrals of the systems vector field, we employ Lie-bracket-based (i.e., derivative-based) averaging.

The brief contains numerous examples and applications, including examples with unknown control directions and experiments with charged particle accelerators. It is intended for theoretical control engineers and mathematicians, and practitioners working in various industrial areas and in robotics.

Presents a unique use of extremum seeking for stabilization of unstable dynamical systems Reinforces the basic principles with extensive use of worked examples Offers a thorough, self-contained analytical study with clear proofs of all results Backs up the theory with detailed experimental, in-hardware results

Autorentext

Alexander Scheinker is with the radio frequency control group at Los Alamos National Laboratory. His research is in dynamical systems and control theory with applications to uncertain, nonlinear, and time-varying systems with a focus on utilizing extremum seeking as feedback control for unknown, open-loop unstable systems. He has been working at the Los Alamos Neutron Science Center linear particle accelerator, developing new algorithms and implementing various control algorithms in hardware. Miroslav Krsti**** ****holds the Alspach endowed chair and is the founding director of the Cymer Center for Control Systems and Dynamics at UC San Diego. He also serves as Associate Vice Chancellor for Research at UCSD. Krstic is Fellow of IEEE, IFAC, ASME, SIAM, and IET (UK), Associate Fellow of AIAA, and foreign member of the Academy of Engineering of Serbia. He has received the PECASE, NSF Career, and ONR Young Investigator awards, the Axelby and Schuck paperprizes, the Chestnut textbook prize, the ASME Nyquist Lecture Prize, and the first UCSD Research Award given to an engineer. Krstic has also been awarded the Springer Visiting Professorship at UC Berkeley, the Distinguished Visiting Fellowship of the Royal Academy of Engineering, the Invitation Fellowship of the Japan Society for the Promotion of Science, and the Honorary Professorships from the Northeastern University (Shenyang), Chongqing University, Donghua University, and Dalian Maritime University, China. Krstic has coauthored eleven books on adaptive, nonlinear, and stochastic control, extremum seeking, control of PDE systems including turbulent flows, and control of delay systems.

Inhalt
Introduction.- Weak Limit Averaging for Studying the Dynamics of Extremum-Seeking-Stabilized Systems.- Minimization of Lyapunov Functions.- Control Affine Systems.- Non-C2 Extremum Seeking.- Bounded Extremum Seeking.- Extremum Seeking for Stabilization of Systems Not Affine in Control.- General Choice of Extremum-Seeking Dithers.- Application Study: Particle Accelerator Tuning.