Back to Search
Start Over
Recurrent neural network-based predefined time control for morphing aircraft with asymmetric time-varying constraints.
- Source :
-
Applied Mathematical Modelling . Nov2024, Vol. 135, p578-600. 23p. - Publication Year :
- 2024
-
Abstract
- • A predefined time control is proposed for morphing aircraft with state constraints. • The unknown disturbances can be estimated faster and more accurately. • The feasibility conditions are not required. • The system state can be limited to preset constraints. • The practical predefined time stability of attitude tracking error is guaranteed. This paper proposes a predefined time control method for morphing aircraft based on an adaptive full-feedback recurrent neural network and a universal barrier function. This method allows morphing aircrafts to track their attitude accurately even under the influence of morphing disturbance, aerodynamic uncertainties and asymmetric time-varying constraints. Specifically, a new universal barrier function is applied to directly perform unconstrained transformations of state variables, which avoids feasibility conditions. Additionally, an adaptive full-feedback recurrent neural network structure is proposed to quickly and accurately approximate additional disturbances and unknown dynamics. Moreover, a backstepping framework is applied to design the predefined time control law, and a command filter is used to prevent the "explosion of complexity" problem. According to stability analyses, the states of the closed-loop system converge within the preset time without violating the state constraints. Finally, the effectiveness of the control algorithm is verified via simulation experiments. [ABSTRACT FROM AUTHOR]
- Subjects :
- *CLOSED loop systems
*RECURRENT neural networks
*TIME-varying networks
Subjects
Details
- Language :
- English
- ISSN :
- 0307904X
- Volume :
- 135
- Database :
- Academic Search Index
- Journal :
- Applied Mathematical Modelling
- Publication Type :
- Academic Journal
- Accession number :
- 179240081
- Full Text :
- https://doi.org/10.1016/j.apm.2024.06.024