论文部分内容阅读
虽然在自动目标识别(ATR)和其他模式识别/分类应用领域有成功地使用神经网络(NN)算法的巨大潜力,但是(如本文所述的)遇到了在反馈控制设计中至今不能严格使用NN的一些值得注意的障碍。本文介绍与控制和估计理论(以及与INS/GPS导航系统中实际的故障检测)有关的涉及NN的若干问题和矛盾的现状。为了促进发展和给出综合展望,我们也介绍应用控制和估计理论中的技术和结果所取得的许多新的值得称赞的NN结果。
Although there is great potential for the successful use of neural network (NN) algorithms in Automatic Target Recognition (ATR) and other pattern recognition / classification applications, (as described herein) have hitherto not been able to strictly use NN in feedback control design Some noteworthy obstacles. This article presents the current status of a number of issues and inconsistencies related to NN related to control and estimation theory (and actual fault detection in INS / GPS navigation systems). In order to foster development and give a synthesis of perspectives, we also present many new commendable NN results achieved by applying techniques and results from applied control and estimation theory.