Aircraft Mass Estimation using Quick Access Recorder Data

Aircraft mass is the most important parameter airliners use to determine how much cargo and fuel they should take for the flight to maximize their profit while keep the total weight under the safety limit. For many practical reasons, airliners can not get the accurate aircraft mass data by weighting every cargo and passenger for each flight. Several studies have proposed methods to estimate aircraft mass based on radar data or ADS-B data. But due to the measurement errors in the data, and also due to uncertainties in aircraft configuration, it is hard to calculate accurate aircraft mass data for every flight. In the paper, the quick access recorder data are used for analysis, parameters that are not available in radar data or ADS-B data can now be used to eliminate parameters with large errors in the level flight point-mass dynamics model. Equations are reformulated as a set of overdetermined linear equations with nonlinearly structured errors in system matrix. The set of equations does not depend on inaccurate parameters like thrust and it does not require accurate knowledge of the aircraft like geometry, aerodynamic coefficients, etc. The set of equations is solved by an improved structured nonlinear total least squares method using Monte Carlo method. The method is applied to 120 real flights of Boeing 777-300ER aircraft, the result shows a good accuracy for some type of flights.

[1]  Joost Ellerbroek,et al.  Aircraft initial mass estimation using Bayesian inference method , 2018 .

[2]  J. Ben Rosen,et al.  Structured Total Least Norm for Nonlinear Problems , 1998, SIAM J. Matrix Anal. Appl..

[3]  Heinz Erzberger,et al.  Adaptive Trajectory Prediction Algorithm for Climbing Flights , 2012 .

[4]  Daniel Raymer.  Aircraft Design: A Conceptual Approach 5e and RDSWin STUDENT , 2012 .

[5]  Joost Ellerbroek,et al.  Modeling and Inferring Aircraft Takeoff Mass from Runway ADS-B Data , 2016 .

[6]  Xu Peiliang,et al.  Overview of Total Least Squares Methods , 2013 .

[7]  J. Ben Rosen,et al.  Total Least Norm Formulation and Solution for Structured Problems , 1996, SIAM J. Matrix Anal. Appl..

[8]  J. Anderson,et al.  Fundamentals of Aerodynamics , 1984 .

[9]  Nicolas Durand,et al.  Learning the aircraft mass and thrust to improve the ground-based trajectory prediction of climbing flights , 2013 .

[10]  Nicolas Durand,et al.  Machine Learning and Mass Estimation Methods for Ground-Based Aircraft Climb Prediction , 2015, IEEE Transactions on Intelligent Transportation Systems.

[11]  Jerry M. Mendel,et al.  The constrained total least squares technique and its applications to harmonic superresolution , 1991, IEEE Trans. Signal Process..

[12]  Kai-Bor Yu,et al.  Total least squares approach for frequency estimation using linear prediction , 1987, IEEE Trans. Acoust. Speech Signal Process..

[13]  R. Kumaresan,et al.  Estimating the parameters of exponentially damped sinusoids and pole-zero modeling in noise , 1982 .

[14]  Hak-Tae Lee,et al.  Closed-Form Takeoff Weight Estimation Model for Air Transportation Simulation , 2010 .

[15]  Joost Ellerbroek,et al.  Bayesian Inference of Aircraft Initial Mass , 2017 .