General methods of estimate fusion applied to spacecraft rendezvous

Techniques for optimally mixing the outputs from a pair of Kalman filters are presented which generalize results previously presented. These techniques are derived under the assumption that the designs of the filters are fixed, and cannot be modified to support fusion requirements. A sufficient condition for using the optimally fused estimates to periodically reinitialize the Kalman filters is presented as well. The results are applied to an optimal spacecraft rendezvous problem, and simulated performance results indicate that use of the optimally fused data leads to significantly improved robustness to initial target vehicle state errors. Two other applications of estimate fusion methods to spacecraft rendezvous are also discussed, state vector differencing and redundancy management.