一种迭代离差差分滤波算法及其应用研究-控制与决策.PDFVIP

绗?6 鍗 绗? 鏈 鎺 鍒 涓 鍐 绛 2011 骞 9 鏈 Vol. 26 No. 9 Control and Decision Sep. 2011 鏂囩珷缂栧彿: 1001-0920 (2011) 09-1425-04 涓€绉嶈凯浠ｇ宸樊鍒嗘护娉㈢畻娉曞強鍏跺簲鐢ㄧ爺绌 1,2 1 绌 闈 , 钄¤繙鍒 (1. 瑗垮畨浜ら€氬ぇ瀛︾數瀛愪笌淇℃伅宸ョ▼瀛﹂櫌锛岃タ瀹?10049 锛?. 瑗垮畨宸ヤ笟澶у璁＄畻鏈虹瀛︿笌宸ョ▼瀛﹂櫌锛岃タ瀹?10032) 鎽 瑕? 閽堝鎵╁睍鍗″皵鏇兼护娉?EKF) 鍜岃凯浠KF 閲忔祴鏇存柊杩囩▼閲囩敤绾挎€у寲璇樊浼犻€? 瀵艰嚧鐘舵€佷及璁＄簿搴﹀亸浣庣殑闂 棰? 灏嗚凯浠ｆ柟娉曘€佺粺璁＄嚎鎬у寲璇樊浼犻€掑拰绂诲樊宸垎婊ゆ尝鍣ㄧ浉缁撳悎, 寤虹珛浜嗕竴绉嶆柊鍨嬭凯浠ｇ宸樊鍒嗘护娉㈡柟娉? 灏嗚鏂规硶 搴旂敤浜庡啀鍏ュ脊閬撶洰鏍囩姸鎬佷及璁? 浠跨湡瀹為獙缁撴灉鏄剧ず, 姝ゆ柟娉曢檷浣庝簡娴嬮噺鏂圭▼鐨勯潪绾挎€у婊ゆ尝鐨勫奖鍝? 鏈夋晥鎻愰珮浜嗙洰鏍 鐨勭姸鎬佷及璁＄簿搴? 鍏抽敭璇岤 鍐嶅叆寮归亾鐩爣鐘舵€佷及璁★紱澶氱淮Stirling 鍐呮彃澶氶」寮忥紱绂诲樊宸垎婊ゆ尝锛涚粺璁＄嚎鎬у寲璇樊浼犳挱 涓浘鍒嗙被鍙耳 TP391 鏂囩尞鏍囪瘑鐮? A Iterated divided difference 铿乴ter and its applications MU Jing 1,2 , CAI Yuan-li1 (1. School of Electronic and Information Engineering锛孹i鈥檃n Jiaotong University 锛孹i鈥檃n 710049 锛孋hina 锛?. School of Computer Science and Engineering锛孹i鈥檃n Technological University 锛孹i鈥檃n 710032 锛孋hina. Correspondent 锛歁U Jing 锛孍-mail 锛歮ujing1977@163. com) Abstract: The linearization error, which is propagated in the measurement update of the extend Kalman 铿乴ter(EKF) and iterated extended Kalman 铿乴ter(IEKF), may somewhat degrade the 铿乴tering performance. Therefore, an iterated divided difference 铿乴ter is proposed, which combines the iterated method and statistically linearization error propagation with the divided difference 铿乴ter. This algorithm is applied to state estimation for reentry ballistic target. Simulation results show that, the proposed method can reduce the in铿倁ence of the measurement nonlinearity, and higher accuracy of state estimation is guaranteed. Key words: state estimation for reentry ballistic target锛沵ult