發(fā)布時(shí)間：2018-09-17 16:38

【摘要】：針對(duì)擴(kuò)展目標(biāo)聯(lián)合估計(jì)運(yùn)動(dòng)狀態(tài)和目標(biāo)外形的問題,提出了一種基于星-凸形隨機(jī)超曲面模型的擴(kuò)展目標(biāo)高斯混合概率密度濾波算法。該算法利用星-凸形隨機(jī)超曲面模型對(duì)量測(cè)的擴(kuò)散程度進(jìn)行建模,同時(shí)利用約束對(duì)目標(biāo)外形參數(shù)進(jìn)行限制。在高斯混合概率假設(shè)密度的框架下,通過對(duì)量測(cè)模型下的量測(cè)似然、新息等參數(shù)的求解和更新遞推實(shí)現(xiàn)擴(kuò)展目標(biāo)的跟蹤。仿真實(shí)驗(yàn)表明,所提算法在保證跟蹤有效性和可行性的同時(shí)提高了對(duì)擴(kuò)展目標(biāo)運(yùn)動(dòng)狀態(tài)和目標(biāo)外形的估計(jì)精度。
[Abstract]:Aiming at the problem of joint estimation of moving state and shape of extended target, a hybrid probability density filtering algorithm for extended target Gao Si based on star-convex random hypersurface model is proposed. The algorithm uses star-convex random hypersurface model to model the diffusivity of the measurement and uses constraints to limit the target shape parameters. In the framework of Gao Si's mixed probability assumption density, the extended target tracking is realized by solving and updating the parameters of measurement likelihood, innovation and other parameters under the measurement model. Simulation results show that the proposed algorithm not only ensures the effectiveness and feasibility of tracking, but also improves the estimation accuracy of the moving state and shape of the extended target.
【作者單位】： 河南工學(xué)院電子通信工程系;新鄉(xiāng)學(xué)院計(jì)算機(jī)與信息工程學(xué)院;
【基金】：河南省高等學(xué)校重點(diǎn)科研項(xiàng)目(14A510025,17B510001)
【分類號(hào)】：TN713
，

本文編號(hào)：2246508