Distributed extended object tracking information filter over sensor networks

This work aims to design a distributed extended object tracking system over realistic network, where both the extent and kinematics are required retain consensus within entire network. To this end, we resort multiplicative error model (MEM) that allows parameters of perpendicular axis-symmetric objects have individual uncertainty. incorporate MEM into information filter (IF) style, use moment-matching technique derive two pair linear models with only additive noise. The separation is merely in fashion, cross-correlation between states preserved as each other's model. As result, closed-form expressions transferred an alternating iteration IFs. With models, centralized IF proposed wherein measurements converted summation innovation parts. Later, under sensor network communication nodes nodes, present IFs through on measurement schemes, respectively. Moreover, prove estimation errors exponentially bounded mean square. benefits testified by numerical experiments comparison state-of-the-art filters literature.