针对被动多传感器机动目标跟踪系统中,由于目标机动性能的不确定以及存在的非线性而导致系统模型与目标实际运动模式难以匹配的问题,提出一种新的自适应曲线模型跟踪算法.该算法通过建立新的方向角模型,设计一种自适应的转弯角速度估计方法,实时计算每个采样时刻目标的切向加速度,以获得与目标实际运动模式相匹配的运动模型,并与扩展卡尔曼滤波相结合,有效提高了被动多传感器下机动目标的跟踪精度. In the passive multi-sensor maneuvering target tracking system, a new adaptive curve model tracking algorithm is proposed due to the uncertainty of the target maneuverability and the existing non-linearity, which makes it difficult to match the system model with the target actual motion mode. By establishing a new steering angle model, an adaptive corner angular velocity estimation method is designed to calculate the tangential acceleration of the target at each sampling instant in real-time to obtain the motion model that matches the actual target motion mode. The proposed model is combined with Extended Kalman Filter Combining effectively improves the tracking accuracy of maneuvering targets under passive multi-sensor.