The single-platform bearings-only track initiation problem is addressed in a multi-target scenario using measurements from a single sonar. The data association problem and initial state estimation problem are formulated jointly by using the maximum likelihood principle. A simulated annealing-like algorithm, which solves the data association and nonlinear least squares estimate iteratively, is proposed to optimize the likelihood function. The computer simulation results show that the algorithm is robust under various experimental conditions.
展开▼