Scheduled Managing Policy for Nonlinear Stochastic Bio-inspired Control Systems via a Robust Impulsive Tracking Control Strategy

• Main Authors: Lin, Ying-Po, 林英博
• Other Authors: Chen, Bor-Sen
• Format: Others
• Language: en_US
• Published: 2014
• Online Access: http://ndltd.ncl.edu.tw/handle/q3k92z

博士 === 國立清華大學 === 電機工程學系 === 102 === The dynamics of bio-inspired control systems mainly rely on the trophic-like interactions among the controlled objects and their predator-like or prey-like control agents. The amounts of the bio-inspired control agents like pesticide, fertilizer or drugs are difficult to be arbitrarily specified during the whole control process, and may be fewer and fewer because they are absorbed by controlled objects like pest, crop or pathogen, or decay with time. Therefore, for some managing purpose or model reference tracking, a proper scheduled managing policy by impulsively introducing bio-inspired control agents, such as periodic irrigation/feeding for crop/livestock or scheduled dosing/treatment for patients, will be necessary. The scheduled managing policy should ensure a robust reference tracking performance even under the interference from random intrinsic fluctuation and uncertain environmental disturbance. In this study, we propose a scheduled managing policy for nonlinear stochastic bio-inspired control systems via a robust impulsive tracking control strategy to attenuate the effects of random intrinsic fluctuations and uncertain environmental noises on desired reference tracking and impulsive control input. To simplify the design procedure without solving a complex Hamilton Jacobi integration inequality (HJII), we combine the global linearization approach with numerical approximation to deal with nonlinearity and integration in HJII so that an equivalent LMIs is proposed for solving this robust impulsive tracking control problem efficiently. Simulation results are also given to verify the efficiency of the impulsive tracking control design of nonlinear stochastic bio-inspired control systems.