Analysis of adaptive neural networks for helicopter flight controls

Leitner, Jesse; Calise, Anthony J.; Prasad, J. V. R.

doi:10.2514/6.1995-3268

Cited by 25 publications

(17 citation statements)

References 1 publication

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Based on the expression in (7) and the subsequent stability analysis, the control input is designed as (13) where are positive definite, diagonal matrices of constant control gains, was defined in (4), and denotes the identity matrix. To simplify the notation in the subsequent stability analysis, the constant auxiliary matrix is defined as (14) where can be separated into diagonal (i.e., ) and off-diagonal (i.e., ) components as (15) After substituting the time derivative of (13) into (7), the following closed-loop error system is obtained:…”

Section: A Error Systemmentioning

confidence: 99%

“…However, several efforts (cf., [7], [14]- [17]) have been developed for the more general problem where the uncertain parameters or the inversion mismatch terms do not satisfy the linear-in-the-parameters assumption (i.e., non-LP). One method to compensate for non-LP uncertainty is to exploit a neural network as an online function approximation method as in [15] and [16]; however, all of these results yield uniformly ultimately bounded stability due to the inherent function reconstruction error. In [14] and [18], bounded aircraft tracking is proven using DI-based control strategies, where neural networks are utilized to compensate for model error.…”

mentioning

confidence: 99%

See 1 more Smart Citation

Asymptotic Tracking for Aircraft via Robust and Adaptive Dynamic Inversion Methods

MacKunis

Patre

Kaiser

et al. 2010

IEEE Trans. Contr. Syst. Technol.

View full text Add to dashboard Cite

Abstract-Two asymptotic tracking controllers are designed in this paper, which combine model reference adaptive control and dynamic inversion methodologies in conjunction with the robust integral of the signum of the error (RISE) technique for output tracking of an aircraft system in the presence of parametric uncertainty and unknown, nonlinear disturbances, which are not linearly parameterizable (non-LP). The control designs are complicated by the fact that the control input is multiplied by an uncertain, non-square matrix. A robust control design is presented first, in which partial knowledge of the aircraft model along with constant feedforward estimates of the unknown input parameters are used with a robust control term to stabilize the system. Motivated by the desire to reduce the need for high-gain feedback, an adaptive extension is then presented, in which feedforward adaptive estimates of the input uncertainty are used. These results show how asymptotic tracking control can be achieved for a nonlinear system in the presence of a non-square input matrix containing parametric uncertainty and nonlinear, non-LP disturbances. Asymptotic output tracking is proven via Lyapunov stability analysis, and high-fidelity simulation results are provided to verify the efficacy of the proposed controllers.

show abstract

Section: A Error Systemmentioning

confidence: 99%

mentioning

confidence: 99%

Asymptotic Tracking for Aircraft via Robust and Adaptive Dynamic Inversion Methods

MacKunis

Patre

Kaiser

et al. 2010

IEEE Trans. Contr. Syst. Technol.

View full text Add to dashboard Cite

show abstract

“…Flight control system design is considered one of the core issues in the development of a fully functional unmanned helicopter, many control system designs have been developed for controlling the dynamic process of miniature helicopter; few among those designs are readily suitable for applying [5]. Several controller design methods are successfully implemented as control technologies such as the adaptive control [6,7], optimal control [8], neural network approach [9,10], the robust and H ∞ control approach [11], and fuzzy logic [12]. However, many of these reported works have one common drawback, they focus merely on the basic miniature helicopter equations of motion without including both the swashplate and mixer mechanism dynamics within miniature helicopter mathematical model, so the efficiency of control system is limited.…”

Section: Introductionmentioning

confidence: 99%

Fuzzy Logic Attitude Control System for a Mini Helicopter Expanded Non Linear Mathematical Model

Al-Shehabi¹

2016

JAEST

View full text Add to dashboard Cite

Full nonlinear mathematical model for a mini helicopter is very complex, inflexible and has big influence on the designed control system performance. Analytical approach for autonomous helicopter modeling including both swash plate and mixer dynamics additional to miniature helicopter equations of motion is introduced. Attitude control system without linearization process is directly designed using fuzzy logic concepts, fuzzy controllers are implemented in both stability and command augmentation loops. Scaling factors are adjusted to minimize the error. The performance of control system for miniature helicopter expanded nonlinear model using fuzzy logic controllers is evaluated through simulation.

show abstract

“…To name a few, in [17], [18], [19], [20], [21], neural networks were fused with ADI based controllers where boundedness of the tracking error was ensured, however the aforementioned studies were unable to prove asymptotic tracking. In [22] and [23], Shin et al developed a tracking controller by fusing robust integral of the signum of the error (RISE) feedback with neural network feedforward terms for a rotorcraft-based UAV and guaranteed semi-global asymptotic tracking.…”

Section: Introductionmentioning

confidence: 99%

A robust adaptive tracking controller for an aircraft with uncertain dynamical terms

Tanyer

Tatlıcıoğlu

Zergeroğlu

2014

IFAC Proceedings Volumes

View full text Add to dashboard Cite

This work presents, the design and the corresponding analysis of a nonlinear controller for an aircraft system subject to uncertainties in the dynamics and additive state-dependent nonlinear disturbance-like terms. Specifically; dynamic inversion technique in conjunction with a robust integral of the signum of the error feedback and an adaptive term is utilized in the overall controller design. Lyapunov based stability analysis techniques are then utilized to prove global asymptotic convergence of the tracking error.

show abstract

Analysis of adaptive neural networks for helicopter flight controls

Cited by 25 publications

References 1 publication

Asymptotic Tracking for Aircraft via Robust and Adaptive Dynamic Inversion Methods

Asymptotic Tracking for Aircraft via Robust and Adaptive Dynamic Inversion Methods

Fuzzy Logic Attitude Control System for a Mini Helicopter Expanded Non Linear Mathematical Model

A robust adaptive tracking controller for an aircraft with uncertain dynamical terms

Contact Info

Product

Resources

About