IEEE Access (Jan 2022)

Stochastic Robust H<sub>&#x221E;</sub> Decentralized Network Formation Tracking Control of Large-Scale Team Satellites via Event-Triggered Mechanism

  • Bor-Sen Chen,
  • Ying-Shuo Ma,
  • Min-Yen Lee

DOI
https://doi.org/10.1109/ACCESS.2022.3180166
Journal volume & issue
Vol. 10
pp. 62011 – 62036

Abstract

Read online

The team formation of large-scale satellites for communication services of future smart cities in the 5G and 6G era is an important research topic currently. In this study, a robust $H_{\infty }$ tracking control with an event-triggered mechanism is proposed for the formation of large-scale satellites for communication tasks. At first, a satellite in the team formation is described by a nonlinear stochastic system with intrinsic random fluctuation, external disturbance, and coupling from other satellites. The proposed team formation tracking control for each satellite needs to consider the effect of network-induced delay and packet-dropout as well as save communication resources via an event-triggered mechanism. Further, the desired team formation of large-scale satellites can be prescribed by a set of reference models with the desired formation shape embedded in their reference inputs. Therefore, the large-scale team formation design problem can be simplified as an independent $H_{\infty }$ network model reference tracking control design problem for each satellite to efficiently attenuate the worst-effect of external disturbance, coupling of other satellites, intrinsic random fluctuation, network-induced delay, and packet-dropout on the model reference tracking performance. In order to avoid solving a nonlinear partial differential Hamillion-Jacobin inequality(HJI) for the $H_{\infty }$ decentralized network formation control for each satellite in the team the formation, the T-S fuzzy model is employed to interpolate several local linearized systems to approximate nonlinear satellite systems. Then the HJI in the $H_{\infty }$ decentralized network team formation design problem can be transformed into a set of linear matrix inequalities (LMIs) which can be easily solved by LMI TOOLBOX in MATLAB. Finally, a simulation example of team formation composed of ten satellites is given to illustrate the design procedure and to validate the proposed method in comparison with other methods.

Keywords