DEVELOPMENT OF AN ONLINE PARAMETER ESTIMATION CAPABILITY FOR AIRCRAFT

摘要

Flight Testing is the best means to determine the flying qualities of aircraft and demonstrate compliance to airworthiness regulations. Furthermore, data recorded in flight can also be used to create mathematical models of the aircraft for further testing and development. These models consist of system parameters known as stability and control derivatives, which are determined from flight and wind tunnel tests by parameter estimation techniques. Aircraft system identification as this field is known can be applied to: create models and data sets for aircraft simulators, design flight control laws for stability augmentation systems, and more recently evaluate UAV's. Incentives to perform such a task in real-time include: developing fault-tolerant aircraft architectures and improved flight test efficiency due to rapid data analysis. This paper addresses the issue related to smoothing and differentiating the necessary data for system identification under the constraints of post-manoeuvre performance. Examples of determining the reduced order models for the SPPO mode of the Cranfield University Jetstream-31 (G-NFLA) and a simulated UAV are presented.