基于交互多模型风险灵敏性秩Kalman滤波的机动目标跟踪算法
首发时间:2022-11-30
摘要:针对复杂环境下机动目标运动模型和量测信息面临着模型、参数和噪声统计特征等不确定性问题,提出一种基于风险灵敏准则(risk sensitive, RS)、秩Kalman滤波(rank Kalman filter, RKF)与交互式多模型(interacting multi-model, IMM)的交互多模型风险灵敏性秩Kalman滤波算法(IMM-RS-RKF)。将最小化风险敏感误差准则引入到RKF中,通过风险敏感因子降低状态估计结果对系统不确定性的敏感性,提高非线性跳变系统的鲁棒性和状态估计精度。将二者引入到IMM中,通过模型交互进而提高目标跟踪系统对多样性运动模型的跟踪精度。机动目标跟踪仿真验证了提出的IMM-RS-RKF算法能够有效地提高机动目标状态估计的鲁棒性和精度。
关键词: 目标跟踪 风险灵敏性 交互多模型 秩Kalman滤波
For information in English, please click here
Maneuvering Target Tracking Algorithm Based on Interactive Multi-Model Risk-Sensitive Rank Kalman Filtering
Abstract:To solve the problem of model uncertainties, parameter uncertainties and noise statistical characteristics uncertainties of maneuvering target motion model and measurement information in the complex environments, an interacting multi-model risk-sensitive rank Kalman filter algorithm (IMM-RS-RKF) is proposed based on the risk-sensitive (RS), rank Kalman filter (RKF) and interacting multi-model (IMM). The minimization risk-sensitive error criterion is introduced into the RKF to reduce the sensitivity of the state estimation results to the system uncertainty by risk-sensitive factors, and to improve the robustness and state estimation accuracy for the nonlinear jump systems. The above method is introduced into the IMM to improve the tracking accuracy of the target tracking system for diverse motion models through model interaction. The maneuvering target tracking simulation verifies that the proposed IMM-RS-RKF algorithm can effectively improve the robustness and accuracy of maneuvering target state estimation.
Keywords: target tracking;risk sensitivity;interacting multi-model ;rank Kalman filtering????
基金:
引用
No.****
动态公开评议
共计0人参与
勘误表
基于交互多模型风险灵敏性秩Kalman滤波的机动目标跟踪算法
评论
全部评论0/1000