This research aims at the analysis and development of filtering and estimation methodologies in the areas of ascent trajectory reconstruction and launch vehicle (LV) modeling. The main objective is the development and validation of a complete trajectory estimation process based on post-flight data applicable to small- or medium-size rockets, as the Italian LV, VEGA. To this end, a suitable mathematical model of the vehicle needs to be derived, which accounts for rigid-body translation and rotation, aerodynamic and thrust forces and moments, structural elastic dynamics modeled using rocket bending modes, inertial coupling effects due to nozzle rotation (tail-wags-dog effect), actuators dynamics and time-varying properties (tabular). The trajectory reconstruction will also allow to investigate the actual scattering of (uncertain) model parameters, such as aeromechanical and bending properties, thus improving the overall knowledge of the system based on available flight data.
Only a few works in the literature specifically address the problem of trajectory reconstruction. Analogously, the problem of identifying and improving the knowledge of uncertain LV model parameters on the basis of flight data is still a mostly unexplored research field.
Indeed, the topic dealt with in this research has been mostly faced in the past by each space agency and industry independently and according to their own specific needs.
Innovative solutions, capable to answer the combined problem of state trajectory reconstruction and LV model identification, are thus appealing and the development of a unified framework would represent a major step towards the Italian space transport industry standardization process to analyze the mission data with perspective to improving the safety of future launches mainly through simulations model improvement, identification of flight anomalies, and the performance evaluation of control systems and algorithms for autonomous guidance. In fact, systematic use of flight data to improve the reliability of the mathematical model used for the synthesis and validation of LV flight control systems may allow for less conservative and more performing solutions.
The primary contribution of this research, whose end-goal is the definition of a unified framework for trajectory analysis and reconstruction for small- and medium-size launch vehicles, is the application of state-of-the-art methods to reconstruct the trajectory of a launcher. To this end, several filtering techniques among which batch filter and IEKF will be investigated. In particular, the latter uses twice the EKF so that the forward filter processes the data starting from the initial time and propagating through all the observed data, while a backward filter propagates the state estimate back to the initial time point. This procedure improves the quality of the estimate because the end point of the forward step has benefited from all available measurement data, but the earlier points in the trajectory have benefited from data recorded only up to that time, therefore a backward pass of the filter improves the estimate. This technique is the most promising in terms of achieved results in the literature [1].
Secondly, methods to improve the knowledge of LV model parameters based on flight data will be investigated. To this end, there are several methods to estimate these parameters such as regression methods, maximum likelihood methods, or frequency-domain methods. For aerospace applications, all of these methods were applied in literature especially for the estimate of the aircraft parameters [8]. The parameters estimation process consists of finding values of unknown model parameters, in an assumed model structure, based on noisy measurements. The principle idea to apply this concept to a launch vehicle is (for example) to use forces and moments equations of the model, parametrized by aerodynamics and propulsive coefficients and estimating the latter starting by the inertial measurement unit (IMU) noisy measurements. At the end of the research work, a complete software framework for trajectory reconstruction and system identification purposes (applied to launch vehicles) will be achieved with the perspective to provide a contribution to the Italian space transport industry.
References
[8] Klein, V., and E. A. Morelli. "Aircraft System Identification: Theory and Practice. American Institute of Aeronautics and Astronautics." Inc, Virginia (2006).