Maximum likelihood estimators for rigid body localization using DOA measurement

Rigid body localization (RBL) not only estimates the position of the target, but also obtains the attitude information of the target. The RBL framework of single base station is studied in three-dimensional space. This framework uses a single base station to measure the direction of arrival (DOA) of signal from small-scale wireless sensor network signal installed on the rigid target surface, and then fuses the DOA measurement with the network topology information, and finally proposes two maximum likelihood estimators (MLE) for RBL purpose. The improved Gauss Newton algorithm is adopted to optimize the MLEs of rotation matrix and translation vector and the three-dimensional position and attitude of the object are estimated. The simulation results show that the proposed MLEs can approach the theoretical Cramer Rao Lower Bound, and have better performance with respect to convergence success rate and computation cost.