【摘要】：為了解決標(biāo)準(zhǔn)的交互式多模型(Interacting Multiple Model,IMM)算法中Markov概率轉(zhuǎn)移矩陣固定不變的問題,結(jié)合容積卡爾曼濾波(Cubature Kalman Filter,CKF)算法,提出了一種Markov概率轉(zhuǎn)移矩陣自適應(yīng)的IMM-CKF算法。該算法引入了一個(gè)Markov矩陣元素的調(diào)整系數(shù),在濾波過程中自適應(yīng)調(diào)整Markov概率轉(zhuǎn)移矩陣的每一個(gè)元素。新算法大幅度提高了匹配模型的概率,降低了非匹配模型的影響,同時(shí)改善了標(biāo)準(zhǔn)IMM算法的濾波效果。最后,通過蒙特卡洛仿真實(shí)驗(yàn)驗(yàn)證了自適應(yīng)IMM-CKF算法的跟蹤效果比IMM-CKF算法更好。
[Abstract]:In order to solve the problem of fixed Markov probability transfer matrix in the standard interactive multi-model (Interacting Multiple Model,IMM algorithm, the volumetric Kalman filter (Cubature Kalman Filter,CKF) algorithm is used to solve the problem. An adaptive IMM-CKF algorithm for Markov probabilistic transition matrix is proposed. The algorithm introduces the adjustment coefficient of a Markov matrix element and adaptively adjusts every element of the Markov probability transfer matrix in the filtering process. The new algorithm greatly improves the probability of the matching model, reduces the influence of the mismatch model, and improves the filtering effect of the standard IMM algorithm. Finally, the Monte Carlo simulation results show that the adaptive IMM-CKF algorithm has better tracking performance than the IMM-CKF algorithm.
【作者單位】： 空軍預(yù)警學(xué)院;
【基金】：學(xué)院科研創(chuàng)新基金重大基礎(chǔ)研究專項(xiàng)課題(No.2014ZDJC0102)
【分類號】：TN713
，

本文編號：2414988