Nonlinear and Linear Unstable Aircraft Parameter Estimations Using Neural Partial Differentiation

Sinha, Manoranjan; Kuttieri, R. A.; Chatterjee, Somenath

doi:10.2514/1.57029

Cited by 23 publications

(15 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As a first case flight data of DHC-2 Beaver aircraft [14] linear model have been considered. The aerodynamic derivatives were obtained by parameter estimation from flight data in a standard condition.…”

Section: B Simulated Unstable Flight Datamentioning

confidence: 99%

Stable and Unstable Aircraft Parameter Estimation in Presence of Noise Using Intelligent Estimation Technique

Roy

Peyada

2016

AIAA Atmospheric Flight Mechanics Conference

View full text Add to dashboard Cite

This paper presents intelligent particle swarm optimization (PSO) based adaptive neuro fuzzy inference system (ANFIS) (Sugeno type) as a new approach to the problem of aerodynamic modeling and parameter estimation for both aerodynamically stable and unstable aircraft in presence of measurement error (sensor noise). A PSO optimizer based neuro fuzzy system capable of predicting generalized force and moment coefficients employing measured motion and control variables only, without the requirement of conventional variables or their time derivatives, have been proposed. Furthermore, it is shown that such a model can be used to extract equivalent stability and control derivatives using both linear and nonlinear kinematic models of a rigid aircraft in presence of uncertainties in the form of instrument or measurement error. Simulated data of an unstable test aircraft, derived from a nonlinear six DOF. simulation, have been utilized to illustrate efficacy of the method. Results have been depicted to highlight the usefulness of the proposed algorithm. Nomenclature, D L C C = Drag and lift coefficients , , l m n C C C = Coefficient of rolling, pitching and yawing moment 0 0 0 , , D L y C C C = pressure coefficient 0 0 0 , , l m n C C C = pressure coefficient C N = Normal force coefficient R = Measurement noise covariance matrix u,v,w = Longitudinal, Lateral, and vertical airspeed components  = Vector of unknown parameters V = Velocity vector Y(k) = Estimated response e  = Elevator deflection  = Angle of attack J = Cost function , ,    = Angles of roll, pitch and yaw a,b,c = Arbitrary coefficients i j O = Output of j th neuron of i th layer sf = Spread factor s = Spread  = Deviation

show abstract

Section: B Simulated Unstable Flight Datamentioning

confidence: 99%

Stable and Unstable Aircraft Parameter Estimation in Presence of Noise Using Intelligent Estimation Technique

Roy

Peyada

2016

AIAA Atmospheric Flight Mechanics Conference

View full text Add to dashboard Cite

show abstract

“…In addition, maximizing the minimum inter-personal distance, weighted sums (Sinha, Kuttieri and Chatterjee, 2013;Ting, Wu and Chou, 2014) and other strategies are applied as the criteria for selecting the exact solutions of individuals.…”

Section: Related Workmentioning

confidence: 99%

Particle Swarm Collaborative Practicability Algorithm Based on Partial Differential Exact Solution

2017

Revista Tecnica De La Facultad De Ingenieria Universidad Del Zu

View full text Add to dashboard Cite

For the particle swarm collaborative practicability algorithm, when extended from single objective to multiobjective problem, the storage and maintenance of the partial differential exact solution sets occurs. And the selection of global and personal exact solutions, balance between exploitation and exploration and other problems also occur. In this paper, the diversity and evolution state of population is evaluated on the distribution and differential entropy, in the new objective space called parallel lattice coordinate system, using the partial differential front end. It applies the information to design the evolutionary strategy, so that the algorithm can consider both the convergence and diversity of approximate partial derivative front end. Also, the concepts of lattice dominance and lattice distance density are introduced to evaluate the personal environmental adaptability of partial derivative exact solution. And the updating methods for external files and selection mechanism for global exact solution are established. Finally, the particle swarm collaborative practicability algorithm is formed based on partial differential exact solution. Experimental results show that: Compared with the other 8 peer-topeer algorithms, this algorithm has demonstrated concentrated significant performance superiority in 6 multiobjective test problems composted of ZDT and DTLZ series on IGD performance index.

show abstract

“…However, this method requires a large number of hidden neurons and is slower than the earlier methods. A similar approach has been reported where a physical insight in the form of partial differentiation approach has been used for estimating the parameters, and has shown satisfactory results 20,21 . However, ANN based methods are dependent on the architecture and training algorithm.…”

Section: Introductionmentioning

confidence: 99%

“…The short-period motion of the same aircraft as described in the last subsection has been generated through simulation of a nonlinear model as presented 21 . The state equation of the nonlinear model is: …”

Section: Nonlinear Unstable Aircraftmentioning

confidence: 99%

Parameter Estimation of Unstable Aircraft using Extreme Learning Machine

Verma¹,

Peyada²

2017

Def. Sc. Jl.

View full text Add to dashboard Cite

The parameter estimation of unstable aircraft using extreme learning machine method is presented. In the past, conventional methods such as output error method, filter error method, equation error method and nonconventional method such as artificial neural-network based methods have been used for aircraft's aerodynamic parameter estimation. Nowadays, a trend of finding an accurate nonlinear function approximation is required to represent the aircraft's equations-of-motion. Such type of nonlinear function approximation is usually achieved using artificial neural-network which is trained with the aircraft input-output flight data using a training algorithm. The accuracy of estimated parameters, which is achieved using the trained network, is highly dependent on the generalisation capability of the network which can be improved using extreme learning machine based network in contrast to artificial neural-network. To estimate the unstable aircraft parameters from the simulated flight data, Gauss-Newton based optimisation method has been used with a predefined aerodynamic model using the trained network. Further, the confidence of the estimated parameters has been shown in comparison to that of the standard parameter estimation methods in terms of the Cramer-Rao bounds.

show abstract

Nonlinear and Linear Unstable Aircraft Parameter Estimations Using Neural Partial Differentiation

Cited by 23 publications

References 21 publications

Stable and Unstable Aircraft Parameter Estimation in Presence of Noise Using Intelligent Estimation Technique

Stable and Unstable Aircraft Parameter Estimation in Presence of Noise Using Intelligent Estimation Technique

Particle Swarm Collaborative Practicability Algorithm Based on Partial Differential Exact Solution

Parameter Estimation of Unstable Aircraft using Extreme Learning Machine

Contact Info

Product

Resources

About