Lateral Aerodynamic Parameters Estimation using Neuro Artificial Bee Colony Fusion Algorithm (NABC)

Abstract

Aerodynamic parameter estimation entails modelling force and moment coefficients as well as computing stability and control derivatives from flight data. This topic has been thoroughly researched utilizing traditional procedures such as output, filter, and equation error methods. Machine learning, such as artificial neural networks, provides an alternate way to these model-based methodologies. This paper proposes a novel estimation technique for aerodynamic parameters of a real aircraft in the presence of system and measurement uncertainty. A fusion between biologically inspired optimization i.e., Artificial Bee Colony (ABC) optimization and widely used Artificial Neural Network (ANN), which mimics the functional unit of the brain, the neuron, has been demonstrated to be novel and a promising method to the challenges of system identification and parameter estimation (sensor noise). The obtained results were compared to Least Square, and Maximum Likelihood Method (MLE), benchmark estimation techniques.