Perpustakaan Digital ITB

In the realm of aircraft development, leveraging flight test data is pivotal for enhanced insights into performance and behaviour. This research revolves around System Identification as a fundamental facet of aircraft development. Investigating the potential of Bayesian Inference in Parameter Estimation, this research integrates Bayesian Inference following model structure determination via the Least Squares method, exclusively for parameter estimation. Employing the Metropolis-Hastings algorithm, the research conducts posterior distribution sampling within the Bayesian Inference framework. Additionally, the research encompasses Uncertainty Quantification, grounded in the analysis of parameter distributions. In sum, this study bridges System Identification, Bayesian Inference, and Uncertainty Quantification, contributing to a comprehensive framework for informed aircraft parameter estimation.