Uncertainty Propagation for Maneuvering Objects in Chaotic Systems

Abstract

State-of-the-art Anthropogenic Space Object (ASO) tracking algorithms cannot robustly maintain custody of objects in cislunar space. One principal challenge is propagating uncertainty through multiple orbit regimes, each with different dynamic signatures and possible chaos. Rapid uncertainty growth when accounting for maneuvers exacerbates the challenges in generating realistic assessments of uncertainty. The goal of this project is to develop a computationally tractable and accurate method of uncertainty propagation for maneuvering objects that accounts for chaotic dynamics and addresses the needs for data association and orbit determination. The University of Texas at Austin will use its unique expertise in efficient orbit state-uncertainty propagation and data fusion to achieve this goal. The project s first objective is to develop efficient algorithms to propagate trajectories and variational equations (i.e., State Transition Matrices (STMs) and State Transition Tensors (STTs)) subject to chaotic dynamics and maneuvers. This includes reducing the computation time of force models and the integration routines required for the propagation of a single reference trajectory and its local variations. The second objective seeks to derive methods for splitting Gaussian Mixture Models (GMMs) and refining sampling distributions based on local and global measures of sensitivity. Local sensitivity may be quantified using the STM-STT, while global indices account for the variations in the local sensitivity over the domain of the prior probability density. When combined, these indices will inform adaptation of the input domain to efficiently propagate uncertainty. Finally, the third objective will formulate a robust, probabilistic data association process for maneuvering ASOs in cislunar space. To this end, we will combine a novel description of a satellite s possible maneuver prole based on stochastic processes, the newly developed sensitivity indices, and efficient orbit propagation routines to generate a predicted density for data association that is consistent with Bayesian multiple target tracking.

Document Details

Document Type
DoD Grant Award
Publication Date
Feb 06, 2025
Source ID
FA95502410297

Entities

People

  • Brandon A. Jones

Organizations

  • Air Force Office of Scientific Research
  • United States Air Force
  • University of Texas at Austin

Tags

Fields of Study

  • Physics

Readers

  • Distributed Systems and Data Platform Development
  • Space Exploration and Orbital Mechanics.
  • Wave Propagation and Nonlinear Chaotic Dynamics.

Technology Areas

  • AI & ML
  • AI & ML - Bayesian Inference
  • AI & ML - Machine Learning Algorithms
  • Space
  • Space - Orbital Debris
  • Space - Space Objects