基于強(qiáng)跟蹤濾波器的多傳感器非線性動態(tài)系統(tǒng)狀態(tài)與參數(shù)聯(lián)合估計(jì)

本文將強(qiáng)跟蹤濾波理論與多傳感器數(shù)據(jù)融合技術(shù)相結(jié)合,提出基于強(qiáng)跟蹤濾波器的多傳感器狀態(tài)與參數(shù)聯(lián)合估計(jì)新算法;對擁有相同采樣率的分布式多傳感器單模型非線性動態(tài)系統(tǒng),應(yīng)用強(qiáng)跟蹤濾波器,得到目標(biāo)狀態(tài)基于全局信息融合估計(jì)結(jié)果,并利用計(jì)算機(jī)仿真結(jié)果對算法的有效性進(jìn)行了驗(yàn)證;這些工作初步解決了Kalman 濾波中由于模型的不確定性而造成估計(jì)誤差值偏大情況下的狀態(tài)融合估計(jì)問題,從而豐富和發(fā)展了多源信息融合理論.
關(guān)鍵詞: 　強(qiáng)跟蹤濾波器; 融合估計(jì); 漸消因子; 動態(tài)系統(tǒng); Kalman 濾波
Abstract : 　By combining the strong tracking filtering theory with data fusion estimation technology ,a new joint state and param2 eter estimation algorithm of multisensor based on strong tracking filter is proposed. For the multisensor and single model nonlinear dy2
namic systems having the same sample rates for every sensor ,the fusion estimate on the basis of global information by use of strong tracking filter is established ,and the effectiveness of the new algorithm is also illustrated by use of an example. These give a primary solution to the fusion estimation problem having bigger errors produced by Kalman filter because of uncertainties of modeling system. This work enriches and develops the information fusion theory.
Key words : 　strong tracking filter ;fusion estimation ;fading factor ;dynamic systems ;Kalman filtering