论文部分内容阅读
针对非线性高斯场景下目标数目未知或随时间变化的机动多目标跟踪问题,提出一种基于交互式多模型的不敏卡尔曼概率假设密度滤波算法.首先,在高斯混合概率假设密度滤波框架下,结合不敏卡尔曼滤波中状态预测和量测更新的实现机理,构建一种不敏卡尔曼概率假设密度滤波器;然后,通过引入交互式多模型方法中状态模型软判决机制,实现对目标机动过程中运动模式不确定的处理;最后,通过理论分析和仿真结果验证了所提出算法的可行性和有效性.
In order to solve this problem, an unstiffened Kalman probability hypothesis density filtering algorithm based on interactive multi-model is proposed for the problem of unknown or time-varying number of targets in a nonlinear Gaussian scene.Firstly, under the Gaussian mixture probability hypothesis density filter framework , A kind of insensitive Kalman probability hypothesis density filter is constructed based on the realization of state prediction and measurement update in unstiffened Kalman filtering. Then, by introducing the soft decision mechanism of state model in the interactive multi-model method, In the end, the feasibility and validity of the proposed algorithm are verified by theoretical analysis and simulation results.