Multisensor Fusion Algorithms for Maneuvering Target Tracking

Utilization of information acquired from a sensor network to improve the tracking accuracy is one of the most important issues in sensor network research. In this paper, two state-vector multisensor fusion algorithms, estimated weights method (EWM) and modified probabilistic neural network (MPNN), using decoupling technique are investigated to handle an arbitrary number of sensors under the assumption that the sensor measurement errors are independent across sensors. Simulation results are presented comparing the performance of the EWM with the MPNN and with the sensor-based decoupled Kalman filtering algorithms