Development of a theoretical basis of robust control for discrete-time linear stochastic systems with uncertain distributions

2016 Fiscal Year Final Research Report

• Project/Area Number: 26820166
• Research Category: Grant-in-Aid for Young Scientists (B)
• Allocation Type: Multi-year Fund
• Research Field: Control engineering/System engineering
• Research Institution: Kyoto University
• Principal Investigator: HOSOE YOHEI 京都大学, 工学研究科, 助教 (50726411)
• Project Period (FY): 2014-04-01 – 2017-03-31
• Keywords: 確率系 / ロバスト制御 / ポリトープ / 不確かさ / 線形行列不等式 / リアプノフ不等式

Outline of Final Research Achievements

We tackled robust control problems for discrete-time linear stochastic systems with dynamics determined by random matrices whose uncertainties can be described with random polytopes. We successfully characterized robust stability of the systems by linear-matrix-inequality-based inequality conditions. Since the inequality conditions included expectations whose exact values are difficult to obtain, we also developed methods for robust stability analysis and controller synthesis in which the expectations are evaluated with samples of random matrices. Then, we further extended the methods so that more advanced robust control such as taking account of stability degree in the synthesis can be achieved.

Free Research Field

制御工学