Instantaneous multisensor angular bias autoregistration

摘要: Unknown alignment biases of sensors a tracking system are estimated by an iterative Kalman filter method. Current measurements corrected for known errors and previously biases. The time reference is updated to produce target state derivative vectors. A Jacobian the dynamics equation determined, which provides observability into sensor bias through gravitational coriolis forces. transition matrix error covariance propagated. When new measurement becomes available, gain vector updated, estimated. vector, measurements, transformed stable space frame use in updating next iteration.