多層神經(jīng)網(wǎng)絡(luò)在跟蹤式卡爾曼濾波器中的應(yīng)用
A IMPROVED TRACKING KALMAN FILTER USING MULTILAYER NEURAL NETWORK
-
摘要: 本文將多層神經(jīng)網(wǎng)絡(luò)引入跟蹤式卡爾曼濾波器中,提高了估計(jì)的精確度。以前的跟蹤式卡爾曼濾波器的估計(jì)精度與目標(biāo)的運(yùn)動(dòng)狀態(tài)有關(guān),當(dāng)目標(biāo)的運(yùn)動(dòng)不能夠用線性狀態(tài)空間模型描述時(shí),其估計(jì)精度將要下降。而多層神經(jīng)網(wǎng)絡(luò)的引入,改善了這一不足。多層神經(jīng)網(wǎng)絡(luò)經(jīng)過訓(xùn)練以后,能夠?qū)柭鼮V波器的結(jié)果進(jìn)行修正。仿真結(jié)果表明,由于多層神經(jīng)網(wǎng)絡(luò)的應(yīng)用,估計(jì)精度顯著提高。
-
關(guān)鍵詞:
- 目標(biāo)跟蹤; 卡爾曼濾波; 多層神經(jīng)網(wǎng)絡(luò)
Abstract: This paper presents a method to improve the estimation accuracy of a tracking Kalman filter (TKF)by using a multilayer neural network(MNN). The estimation accuracy of the TKF is degraded due to the uncertainties that cannot be expressed by the linear state-space model proposed in the literature. This fault is overcome due to the use of MNN. The results of the TKF can be modified by the treated MNN. Simulation results show that the estimation accuracy is much improved by using the MNN. -
Singer R A. Estimating optimal tracking filter performance for manned maneuvering targets IEEE Trans[J].on Aerosp. Electron. Syst.1970, AES-6(4):473-483[2]Singer R A, Behnke K W. Real-time tracking filter evolution and selection for tracking application[3]IEEE Trans. On Aerosp. Electron. Syst., 1970, AES-7(1): 100-110[4]Rumelhart D E, Hilton G E, Williams R J. Learning Internal Representation by Error Propagation. In Parallel Distributed Processing, 1986, Chapter 8, 319-362.[5]Cloutier J R, Evers J H, Feeley J J. Assessment of air-to-air missile guidance and control technology[J].IEEE Cont. Syst. Mag.1989, 9(6):27-34[6]Funahashi K. On the approximate realization of continuous mapping by neural networks. Neural Networks 1989, 2(3): 184-192. -
計(jì)量
- 文章訪問數(shù): 2044
- HTML全文瀏覽量: 189
- PDF下載量: 365
- 被引次數(shù): 0
下載:
