Tracking maneuvering spacecraft with filter-through approaches using interacting multiple models

When spacecraft maneuver unknowingly, traditional orbit determination filters diverge. This paper postulates using Interacting Multiple Models and covariance inflation methods to filter-through unknown maneuvers. This work proposes accurate methods for real-time tracking of a non-cooperative, maneuvering spacecraft and compares the performance to traditional techniques. Results show that a filter-through Interacting Multiple Model orbit determination filter can converge on a post-maneuver orbit in real-time with similar performance to off-line Initial Orbit Determination approaches.