Ryan Russell (University of Texas at Austin), Nitin Arora (University of Texas at Austin), Vivek Vittaldev (University of Texas at Austin), David Gaylor (Emergent Space Technologies, Inc.), Jessica Anderson (Emergent Space Technologies, Inc.)
Keywords: SSA
Abstract:
Recent improvements in atmospheric density modeling now provide more confidence in spacecraft ballistic coefficient (BC) estimations, which were previously corrupted by large errors in density. Without attitude knowledge, forecasting the true BC for accurate future state and uncertainty predictions remains elusive. In this paper, our objective is to improve this predictive capability for ballistic coefficients for Resident Space Objects (RSOs), thus improving the existing drag models and associated accuracy of the U.S. Space Object Catalog. To work towards this goal we implement a two-pronged strategy that includes elements of time series analysis and physics based simulations. State-of-the-art empirical time series prediction methods are applied and tested on BC time series in the context of both simulation data and actual data provided by the Air Force. An archive of simulated BC data is generated using custom 6DOF high fidelity simulations for RSOs using plate models for shapes. The simulator includes force and torque perturbations due to the nonspherical Earth, third-body perturbations, SRP, and atmospheric drag. The simulated BC profiles demonstrate significant variation over short time spans (due primarily to varying frontal areas), providing motivation to improve future BC estimation strategies. The 6DOF modeling is intended to provide a physics-based BC data set to complement the BC data set provided by the AF. For the black-box time series algorithms, a variety of approaches are considered, whereas two prediction models showed the most promising performance: a multi-tone harmonic model and an autoregressive (AR) model. Both the multi-tone harmonic model and the AR model are subjected to multiple levels of optimizations resulting in highly optimized final models that are tuned specifically with the 205 BC time series provided by the AF. Two versions of the AR model are developed based on the model prediction methodology. The second version of the AR model performs approximately as well as the optimized multi-tone model. The proposed algorithms automatically select the fitting function and duration of the fit span tuned for the best prediction performance based on past known data. The results demonstrate the ability to robustly and automatically fit real past BC data in order to predict forward for 1-10 days with improved accuracy. These improvements are mapped to position error predictions for typical satellites, demonstrating the utility of such predictions in the context of conjunction analysis and other catalog applications. The improved performance of the time series prediction algorithms applied to the physics-based simulated data suggests that the actual estimated BCs include other signatures.
Date of Conference: September 11-14, 2012
Track: Poster