Mitigation of the Impact of Sensing Noise on the Precise Formation Flying Control Problem

Faculty contact: Jonathan How
Department of Aeronautics and Astronautics
Aerospace Controls Laboratory
Massachusetts Institute of Technology

Research funded by NASA Cooperative Agreement NCC5-729

Research objective:
Understand and mitigate the effects of navigation errors on spacecraft formation flying missions.

• Analysis of the impact of the velocity errors on spacecraft
  formation flying missions (general orbit configurations)
• Investigate the fundamental performance limits of CDGPS
  estimators, and identify estimator changes that might improve
  upon the accuracy of the current velocity estimates
• Identify augmentation approaches that might be efficiently added to
  the spacecraft to improve velocity estimates (onboard and/or
  ground based)
• Extend current control techniques to further enhance their ability
  to achieve fuel efficient performance even with sensing errors,
  differential drag disturbances, and gravity perturbations.

Focus on:

• Navigation filter parameters and their effect on semimajor axis knowledge
• Analysis of the effects of spacecraft relative navigation errors on
  closed loop control in widely spaced, highly elliptical orbits
• Improving optimized control algorithms by incorporating precise
  knowledge of expected noise characteristics



 

Results in Phase 1:

• Examined effects of navigation filter properties on semimajor axis estimation
  and formation flying control
• Linearized analysis of velocity noise with position measurements ( Paper )
• Extended linearized analysis to include velocity measurements.
  Examined discrete and nonlinear effects. ( Paper )
• Developed linearized dynamics based on GVEs and embedded those
  dynamics in a model predictive controller.
• Development of GVE-based dynamics; demonstration of large baseline
  formation flying. ( Paper )
• Extended GVE-based dynamics to include the effects of
  differential J2 disturbances. ( Paper )
• Demonstrated analytic performance prediction for formation flying mission
  using a model predictive control algorithm. Showed benefits by conducting
  trade studies of control parameters.( Paper )

Recent Results in Phase 2:

• Developed optimized approach to finding J2 invariant relative orbits.
  The method balances fuel use minimization , drift minimization, and template
  matching.
• Verified and implemented a new approach for creating J2-modified discrete
  input matrices.