Robustness and cognition in stabilization problem of dynamical systems based on asymptotic methods

International Conference on Information Technologies in Business and Industry 2016

计算机科学;经济学;工业技术

Dubovik, S.A.^1 ; Kabanov, A.A.^1

Sevastopol State University, 33, Universitetskaya Str., Sevastopol

299053, Russia^1

关键词: Asymptotic method;    Controlled process;    Prediction of critical state;    Prediction systems;    Stabilization problems;    Supervisory control;    Uncertain condition;    Worst case scenario;   

Others  :  https://iopscience.iop.org/article/10.1088/1742-6596/803/1/012035/pdf
DOI  :  10.1088/1742-6596/803/1/012035

来源: IOP

【 摘 要 】

The problem of synthesis of stabilizing systems based on principles of cognitive (logical-dynamic) control for mobile objects used under uncertain conditions is considered. This direction in control theory is based on the principles of guaranteeing robust synthesis focused on worst-case scenarios of the controlled process. The guaranteeing approach is able to provide functioning of the system with the required quality and reliability only at sufficiently low disturbances and in the absence of large deviations from some regular features of the controlled process. The main tool for the analysis of large deviations and prediction of critical states here is the action functional. After the forecast is built, the choice of anti-crisis control is the supervisory control problem that optimizes the control system in a normal mode and prevents escape of the controlled process in critical states. An essential aspect of the approach presented here is the presence of a two-level (logical-dynamic) control: the input data are used not only for generating of synthesized feedback (local robust synthesis) in advance (off-line), but also to make decisions about the current (on-line) quality of stabilization in the global sense. An example of using the presented approach for the problem of development of the ship tilting prediction system is considered.