基于QPF的无源传感器目标跟踪算法

Passive Sensor Target Tracking Algorithm Based on QPF

在无源传感器目标跟踪系统的研究中,在双红外传感器组成的无源传感器目标跟踪系统模型中,传感器提供的仅是目标的角度信息,导致量测值与目标状态之间存在较强的非线性关系,而传统跟踪算法在解决非线性问题时具有局限性,为提高目标跟踪精度,提出一种基于积分粒子滤波(QPF)的无源传感器目标跟踪算法,在不受非线性、非高斯问题限制的粒子滤波的基础上,从改进粒子滤波重要性函数的角度入手,利用积分卡尔曼滤波(QKF)将当前最新量测考虑在内,构造出粒子滤波的重要性函数,使得改进后的重要性函数更加贴近真实后验分布,从而有效遏制了粒子退化现象。仿真结果表明,改进算法提高了跟踪精度,较好地解决了无源传感器对目标的非线性跟踪优化问题。

Only angle information of a target is provided by sensors in a passive sensor tracking system with dual infrared sensors. It inevitably leads to a strong nonlinear relationship between the measured value and the state value. Traditional tracking algorithms cannot effectively filter the nonlinearity. In order to improve the target tracking preci- sion, a passive sensor target tracking algorithm is presented based on the Quadrature Particle Filter （QPF）, which is not limited by nonlinear and non - Gauss problems. From the perspective of the improved importance function, it takes the up to date measurement into account by means of the Quadrature Kalman Filter （ QKF）, and makes the im- portance function closer to the true posterior. The computer simulation results indicate that the proposed algorithm can improve tracking accuracy and solve the nonlinear tracking problem effectively.