A MPF based distributed method for mobile robot and wireless sensor nodes simultaneous localization

A MPF based distributed method for mobile robot and wireless sensor nodes simultaneous localization is proposed to solve the problem that localization based on the simple particle filter suffers from severe sample degeneracy. In the proposed method, the proposal distribution of each particle is estimated by UKF, and the observations between wireless sensor nodes are used to estimate a sensor node´s location. Simulation results show that the proposed method can increase the sample variety and reduce sample degeneracy, and is much more accurate, compared with distributed localization based on the simple particle filter. Moreover, the estimation accuracy of the proposed method is still good with few sample. Therefore, it is feasible to improve the computational efficiency by reducing the number of particles.