Robust Aerodynamic Model Identification: A New Method for Aircraft System Identification In the Presence of General Dynamic Model Deficiencies

Moszczynski, Gregory J.; Leung, Jordan; Grant, Peter R.

doi:10.2514/6.2019-0433

Cited by 4 publications

(7 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Hence, the linear least-squares method may be incapable of accurately modeling the nonlinear relationship between the MFE variations and structural failure degrees. Assuming the availability of proper system identification methods [39,66] to estimate the aerodynamic derivatives, it is anticipating that the nonlinear regression method and specifically neural networks with their significant learning capabilities would be able to perform the aforementioned modeling in the case of structural damages. This can be studied in future researches as the specific numerical methods and wind tunnel tests required for the evaluation of the force and moment non-dimensional derivatives at various structural failure degrees are beyond the scope and available resources of this study.…”

Section: Fig 14 Two-layer Feedforward Neural Network Architecturementioning

confidence: 99%

Investigating Impaired Aircraft's Flight Envelope Variation Predictability Using Least-Squares Regression Analysis

Norouzi

Kosari

Sabour

2020

Journal of Aerospace Information Systems

View full text Add to dashboard Cite

Aircraft failures alter the aircraft dynamics and cause maneuvering flight envelope to change. Such envelope variations are nonlinear and generally unpredictable by the pilot as they are governed by the aircraft complex dynamics. Hence, in order to prevent in-flight Loss of Control it is crucial to practically predict the impaired aircraft's flight envelope variation due to any a-priori unknown failure degree. This paper investigates the predictability of the number of trim points within the maneuvering flight envelope and its centroid using both linear and nonlinear least-squares estimation methods. To do so, various polynomial models and nonlinear models based on hyperbolic tangent function are developed and compared which incorporate the influencing factors on the envelope variations as the inputs and estimate the centroid and the number of trim points of the maneuvering flight envelope at any intended failure degree. Results indicate that both the polynomial and hyperbolic tangent function-based models are capable of predicting the impaired fight envelope variation with good precision. Furthermore, it is shown that the regression equation of the best polynomial fit enables direct assessment of the impaired aircraft's flight envelope contraction and displacement sensitivity to the specific parameters characterizing aircraft failure and flight condition.Nomenclature V = total airspeed, knot ,  = angle of attack, sideslip angle, respectively, deg p, q, r = angular velocity components (roll rate, pitch rate, yaw rate, respectively), deg/s 1 Ph.D. Candidate, Faculty of New Sciences and Technologies; ramin.norouzi@ut.ac.ir. Student Member AIAA. 2 Associate Professor, Faculty of New Sciences and Technologies; kosari_a@ut.ac.ir. 3 Assistant Professor, Faculty of New Sciences and Technologies; sabourmh@ut.ac.ir. , , ,  = Euler angles (roll, pitch, yaw, respectively), flight path angle, deg , = state vector, control vector, respectively , , , ℎ = deflection angles (elevator, aileron, rudder, respectively), deg, throttle setting (%) ∈ [0, 1] , �, = fitted model's residual, outcome, parameters (coefficients) vectors, respectively , = predictor vector, model parameters (coefficients) vector, respectively = total degree of polynomial , ℍ, , = gradient, Hessian, Jacobian of the objective function , identity matrix, respectively = Newton's method search direction = Marquardt parameter = hyperbolic tangent function 1 , 2 = total number of neurons in the first (hidden) layer, second (last) layer, respectively = backpropagation training iteration ℒ , ℒ , ℒ , ℒ, = (net output, net input, bias) of ℒ ℎ neuron in layer , respectively ℒ,= weight's element of ℒ ℎ neuron in layer due to ℎ neuron of the previous layer .̂ = standard backpropagation sensitivity, Marquardt sensitivity, respectively ℰ, = contribution of the ℰ ℎ neuron of the last (second) layer to the error of the ℎ training sample = statistical expected value of the ℎ input factor Subscripts , = total number of training samples, total number of model parameters, res...

show abstract

Section: Fig 14 Two-layer Feedforward Neural Network Architecturementioning

confidence: 99%

Investigating Impaired Aircraft's Flight Envelope Variation Predictability Using Least-Squares Regression Analysis

Norouzi

Kosari

Sabour

2020

Journal of Aerospace Information Systems

View full text Add to dashboard Cite

show abstract

“…The advancement in sensor technology has made the fabrication of micro-electro-sensor systems possible, which helps in logging the flight data acquired while performing system identification manoeuvres, even in small UAVs. Equation error methods (EEM), output error methods (OEM) [14][15][16], filter error methods (FEM) [17][18][19] and Artificial Intelligence-(AI) based methods [20][21][22][23][24] are primary aerodynamic parameter estimation methods. The least square cost function-based EEM has been touted as a promising alternative for a rapid parameter estimation technique because of its computational simplicity.…”

Section: Introductionmentioning

confidence: 99%

Parametric model identification of delta wing UAVs using filter error method augmented with particle swarm optimisation

Kumar

Saderla

et al. 2023

Aeronaut. j.

View full text Add to dashboard Cite

From arsenal delivery to rescue missions, unmanned aerial vehicles (UAVs) are playing a crucial role in various fields, which brings the need for continuous evolution of system identification techniques to develop sophisticated mathematical models for effective flight control. In this paper, a novel parameter estimation technique based on filter error method (FEM) augmented with particle swarm optimisation (PSO) is developed and implemented to estimate the longitudinal and lateral-directional aerodynamic, stability and control derivatives of fixed-wing UAVs. The FEM used in the estimation technique is based on the steady-state extended Kalman filter, where the maximum likelihood cost function is minimised separately using a randomised solution search algorithm, PSO and the proposed method is termed FEM-PSO. A sufficient number of compatible flight data sets were generated using two cropped delta wing UAVs, namely CDFP and CDRW, which are used to analyse the applicability of the proposed estimation method. A comparison has been made between the parameter estimates obtained using the proposed method and the computationally intensive conventional FEM. It is observed that most of the FEM-PSO estimates are consistent with wind tunnel and conventional FEM estimates. It is also noticed that estimates of crucial aerodynamic derivatives ${C_{{L_\alpha }}},\;{C_{{m_\alpha }}},\;{C_{{Y_\beta }}},\;{C_{{l_\beta }}}$ and ${C_{{n_\beta }}}$ obtained using FEM-PSO are having relative offsets of 2.5%, 1.5%, 6.5%, 3.4% and 7.6% w.r.t. wind tunnel values for CDFP, and 1.4%, 1.9%, 0.1%, 9.6% and 7.5% w.r.t. wind tunnel values for CDRW. Despite having slightly higher Cramer-Rao Lower Bounds of estimated aerodynamic derivatives using the FEM-PSO method, the simulated responses have a relative error of less than 0.10% w.r.t. measured flight data. A proof-of-match exercise is also conducted to ascertain the efficacy of the estimates obtained using the proposed method. The degree of effectiveness of the FEM-PSO method is comparable with conventional FEM.

show abstract

“…In Bayoğlu and Nalci (2016) an adaptive KF is applied to APE for a supersonic missile having rapid speed variations. A two-step methodology, studied in Moszczynski, Leung, and Grant (2019), in which an adaptive maximum aposteriori trajectory estimation scheme is adopted for accurate dynamic model identification. Two concepts namely Estimation After Modeling (EAM) and Estimation Before Modeling (EBM) are introduced for the first time by (Mohammadi & Massoumnia, 2000).…”

Section: Introductionmentioning

confidence: 99%

“…Classic filters are more convenient but they have some limitations: Kalman filter is optimal only for linear systems with Gaussian noise. EKF and UKF methods are used for nonlinear systems, but the problem associated with them is greatly attributed to their needed approximations in linearization process (Moszczynski, Leung, & Grant, 2019).…”

Section: Introductionmentioning

confidence: 99%

New Heuristic Algorithms for Rolling Air Frame Aerodynamic Parameters Estimation

Mohamad¹,

Karimi²,

Naderi³

2020

Aviation

View full text Add to dashboard Cite

In this research, based on heuristic optimization algorithms, three new strategies are developed for Aerodynamic Parameters Estimation (APE) of one pair ON-OFF actuator rolling airframe. In the 1st method namely EAM-PSO the aerodynamic parameters are directly estimated. While, the next two algorithms called EBM-PSO and SEBM-PSO are two-step strategies. In the 1st step the aerodynamic forces and moments are estimated, then after passing through a designed smoothing filter, in the 2nd step aerodynamic parameters are estimated. In EBM-PSO all the aerodynamic parameters are estimated at once by solving one optimization problem. In SEBM-PSO the APE is converted to solve four separate optimization problems. A modified particle swarm optimization algorithm is developed and used in estimation process. The performance of proposed algorithms is compared with that of state of the art algorithm EKF. The simulation results show that SEBM-PSO and EBM-PSO are better than EAM-PSO in term of accuracy and run time.

show abstract

Robust Aerodynamic Model Identification: A New Method for Aircraft System Identification In the Presence of General Dynamic Model Deficiencies

Cited by 4 publications

References 32 publications

Investigating Impaired Aircraft's Flight Envelope Variation Predictability Using Least-Squares Regression Analysis

Investigating Impaired Aircraft's Flight Envelope Variation Predictability Using Least-Squares Regression Analysis

Parametric model identification of delta wing UAVs using filter error method augmented with particle swarm optimisation

New Heuristic Algorithms for Rolling Air Frame Aerodynamic Parameters Estimation

Contact Info

Product

Resources

About