Emily Gerber, L3Harris; Thomas Kelecy, L3Harris; Jason Balke, L3Harris
Keywords: Object Characterization, Functional Data Analysis, Data Fusion
Abstract:
The value of techniques which might provide a more relevant characterization, and hence improved information on dynamic and physical attributes of non-resolved debris objects, will support improved attribution and custody of the increasing population of orbital debris. Previous work has proposed a taxonomical approach to characterizing orbital debris. Others have proposed characterization techniques which move away from using time and frequency domain analyses of photometric data samples using information theoretic and functional data analysis (FDA). This approach showed some success in characterizing the physical and dynamic attributes of space objects from non-resolved photometric observations. Though promising, the results also revealed shortcomings where ambiguous characterization of the states resulted due to limited information content when only photometric data were available. Subsequent work demonstrated a remarkable reduction in the ambiguity of characterized states when multiple data types were combined in the characterization step of the FDA process. Inclusion of photometric, tracking, Long Wave Infra-Red (LWIR) and Radio Frequency (RF) measurement types resulted in very reliable characterization of active, passive, dormant and transitioning satellite states.
Other recent research has also demonstrated the viability of collecting spectroscopic measurements to support orbital debris characterization. In this work we continue along taxonomical lines and apply the FDA approach to characterizing orbital debris attributes that include simulated photometric, dynamic, thermal and spectroscopic attributes to selected classes of debris which include a tumbling rocket-body, defunct (uncontrolled) satellites of different types, and typical debris parts such as solar panel material, Multi-Layer Insulation (MLI), aluminum panels, etc. Simulated spectroscopic measurements will also be included in the FDA characterization to evaluate the value of including that data type in the FDA characterization. The analysis will then be extended to include real photometric and astrometric data on active and passive satellites. Additional characterization techniques will be explored and discussed for the different measurement phenomena. Information gain in the characterization when including specific combinations of measurement and analysis types are explored.
Date of Conference: September 14-17, 2021
Track: Non-Resolved Object Characterization