Instantaneous multisensor angular bias autoregistration

Abstract

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