Gaussian mixture PHD filter for multi-sensor multi-target tracking with registration errors

This paper studies the problem of multi-sensor multi-target tracking with registration errors in the formulation of random finite sets. The probability hypothesis density (PHD) recursion is applied by introducing the dynamics of the translational measurement bias into the associated intensity functions. Under the linear Gaussian assumptions on the bias dynamics, the Gaussian mixture implementation is used to give closed-form expressions. As the target state and the translational measurement bias are coupled through the likelihood in the update step, a two-stage Kalman filter is adopted to approximate the tractable form, which leads to a substantial reduction in computational complexity. Two numerical examples are provided to verify the effectiveness of the proposed filter.

► A distributed UKF for nonlinear systems with Gaussian mixture noise is proposed.
► The Gaussian mixture weight is computed by introducing the natural logarithm transformation.
► The proposed filter is robust to sensor failure and scalable.
► Results show the effectiveness of the proposed filter compared to the centralized version.