The unknown maneuver is a key restriction for the real-time spacecraft tracking. For the modeling residual by dynamic of orbits, a compensative modeling method is developed according to the idea of “current” statistic. Then an orbit filter is established based on the new model and the unscented filter. By utilizing a non-mean value and time correlated stochastic process for the maneuvering acceleration, the new model can achieve an online auto-adaptivity for orbit acceleration with different maneuvers mode. Therefore, it can obtain a continuous and accurately estimated orbit for orbit with unknown maneuver. Simulation results for space target with series of random pulsar and continuous maneuvers validated the feasibility and efficiency of the new tracking filter.
[1]
Jun Yang,et al.
Research on Maneuvering Target Tracking Algorithm with Bearing Only Measurements
,
2012
.
[2]
X. R. Li,et al.
Survey of maneuvering target tracking. Part I. Dynamic models
,
2003
.
[3]
X. Rong Li,et al.
A Survey of Maneuvering Target Tracking—Part IV: Decision-Based Methods
,
2002
.
[4]
X. R. Li,et al.
Survey of maneuvering target tracking: II. Ballistic target models
,
2001
.
[5]
Jeffrey K. Uhlmann,et al.
New extension of the Kalman filter to nonlinear systems
,
1997,
Defense, Security, and Sensing.