Robust estimation of a maneuvering target from multiple unmanned air vehicles' measurements

When multiple UAVs collaborate to track a maneuvering target, their position measurement sensors are sometimes corrupted by noise biases (e.g. sensor drifting). In this case, the zero-mean noise assumption of the Kalman filter is therefore violated and the desired optimal estimate will not be guaranteed. In this paper, an H-infinity filter is utilized to estimate the position of the maneuvering target to compensate for non-zero-mean noise. Furthermore, the constrained H-infinity filter is shown to be superior to the Kalman filter.