Unscented particle filter for α-Jerk model with colored noise

As one of the most critical issues for target track, α-Jerk model is an effective maneuvering target track model. Colored noise always exists in the measurement process of the tracking target system which usually influences the tracking performance. A novel maneuvering target track approach based on an unscented particle filter is derived. The unscented particle filter (UPF), which uses the unscented Kalman filter (UKF) to produce the importance density function, is a modified particle filter (PF). This article uses the unscented particle filter to deal with the colored noise on α-Jerk model combined with state variable extension thought. Simulation results indicate that the unscented particle filter has a higher estimate accuracy and overcomes the particle degradation phenomenon effectively at the cost of a little more time consumption.