Skip to main content Accessibility help

Flight performance monitoring with optimal filtering applications

  • V. A. Deo (a1), F. Silvestre (a2) and M. Morales (a3)


This work presents an alternative methodology for monitoring flight performance during airline operations using the available inboard instrumentation system. This method tries to reduce the disadvantages of the traditional specific range monitoring technique where instrumentation noise and cruise stabilisation conditions affect the quality of the performance monitoring results. The proposed method consists of using an unscented Kalman filter for aircraft performance identification using Newton’s flight dynamic equations in the body X, Y and Z axis. The use of the filtering technique reduces the effect of instrumentation and process noise, enhancing the reliability of the performance results. Besides the better quality of the monitoring process, using the proposed technique, additional results that are not possible to predict with the specific range method are identified during the filtering process. An example of these possible filtered results that show the advantages of this proposed methodology are the aircraft fuel flow offsets, as predicted in the specific range method, but also other important aircraft performance parameters as the aircraft lift and drag coefficients (CL and CD), sideslip angle (β) and wind speeds, giving the operator a deeper understanding of its aircraft operational status and the possibility to link the operational monitoring results to aircraft maintenance scheduling. This work brings a cruise stabilisation example where the selected performance monitoring parameters such as fuel flow factors, lift and drag bias, winds and sideslip angle are identified using only the inboard instrumentation such as the GPS/inertial sensors, a calibrated anemometric system and the angle-of-attack vanes relating each flight condition to a specific aircraft performance monitoring result. The results show that the proposed method captures the performance parameters by the use of the Kalman filter without the need of a strict stabilisation phase as it is recommended in the traditional specific range method, giving operators better flexibility when analysing and monitoring fleet performance.


Corresponding author


Hide All
[1]Speyer, J. Auditing aircraft cruise performance in airline revenue service, 7th Performance and Operations Conference, 1992.
[2]Chu, E., Gorinevsky, D. and Boyd, S. Detecting aircraft performance anomalies from cruise flight data, AIAA Infotech Aerospace 2010, American Institute of Aeronautics and Astronautics, April 2010.
[3]Krajcek, K., Nikolic, D. and Domitrovic, A. Aircraft performance monitoring from flight data, Tehnicki vjesnik – Technical Gazette, 2015, 22, (5).
[4]Embraer 170/175/190/195 Airplane Performance Monitoring, May 2011.
[5]Flight Operations Support and Line Assistance, Airbus, Getting Hands On Experience Aerodynamic Deteriorations, 2001.
[6]Airbus Flight Test Guide, May 2013.
[7]Kalman, R. E. A new approach to linear filtering and prediction problems, ASME Journal of Basic Engineering, 1960.
[8]Wan, E. A. and van der Merwe, R. The Unscented Kalman Filter for Nonlinear Estimation, Technical report, Oregon Graduate Institute of Science & Technology, 2000.
[9]Morelli, E. A. Estimating noise characteristics from flight test data using optimal Fourier smoothing, J. Aircr., 1995, 32, (4), pp 689695.
[10]Akhlaghi, S., Zhou, N. and Huang, Z. Adaptive adjustment of noise covariance in Kalman filter for dynamic state estimation, Technical report,Pacific Northwest National Laboratory, February 2017.
[11]Shyam Mohan, M., Naik, N., Gemson, R. and Ananthasayanam, M. Introduction to the Kalman filter and tuning its statistics for near optimal estimates and Cramer Rao Bound, Technical report, Indian Institute of Technology, February 2015.


Flight performance monitoring with optimal filtering applications

  • V. A. Deo (a1), F. Silvestre (a2) and M. Morales (a3)


Full text views

Total number of HTML views: 0
Total number of PDF views: 0 *
Loading metrics...

Abstract views

Total abstract views: 0 *
Loading metrics...

* Views captured on Cambridge Core between <date>. This data will be updated every 24 hours.

Usage data cannot currently be displayed