A maneuvering target tracking algorithm is proposed to overcome the defects of poor filtering precision while using unscented Kalman filter (UKF). The method combines the merits of UKF and adaptive neuro-fuzzy inference system (ANFIS). ANFIS is used to adjust system noise covariance matrix in target tracking system. Fuzzy inference, neural networks and UKF are integrated effectively. The proposed method is applied to the simulation of radar target tracking. The simulation results show that the proposed method has advantages in higher precision, faster convergence, and stronger ability to track maneuvering targets.