论文部分内容阅读
一种基于Kalman和扩展Kalman滤波器的相互作用多模型(IMM)方法可以减小模型的不确定性,但无法消除由于噪声相关引起的状态偏差的弱点。为了提高目标状态估计的精度,把IMM和一种带多重渐消因子的扩展Kalman滤波器(SMFEKF)相结合,提出了一种具有相关噪声的混合随机模型的机动目标跟踪方法。这种方法引入了一个多重渐消因子,当输出残差发生变化时,动态调节增益和系统噪声水平,使输出残差近似正交,从而抑制了相关噪声的影响,适应目标的状态变化。理论分析和仿真实验表明了这种算法的有效性和可行性。
An interaction multi-model (IMM) method based on Kalman and extended Kalman filter can reduce the uncertainty of the model, but can not eliminate the weakness of the state deviation caused by the noise correlation. In order to improve the accuracy of target state estimation, an IMM is combined with an extended Kalman filter with multiple fading factors (SMFEKF) to propose a maneuvering target tracking method based on a hybrid stochastic model with correlated noise. This method introduces a multiple fading factor. When the output residuals change, the gain and the system noise level are dynamically adjusted, and the output residuals are approximately orthogonal, so as to suppress the influence of the relevant noise and adapt to the state change of the target. Theoretical analysis and simulation experiments show the effectiveness and feasibility of this algorithm.