面向多时延多频率测量信息的预测显示算法

发布时间：2018-07-01 17:46

本文选题：预测 + 遥操作　；参考：《机器人》2017年05期

【摘要】：提出了一套能处理多时延、多频率的测量信息和数据缺失问题的预测显示算法,从而充分利用测量信息和提高预测精度.首先通过把部分下传通道时延"移动"到上传通道,把预测显示问题转化为一个标准的滤波问题和一个预测问题,并采用"观测器-预测器"的串联结构解决这2个问题.在观测器中通过重组织新息分析法把多时延的测量信息重组成不包含时延的形式,并借助标记函数来处理异频和缺失的测量信息.仿真实验结果证明了本文预测显示算法的有效性.
[Abstract]:A set of prediction and display algorithms which can deal with the problem of missing measurement information and data with multiple time-delay and multi-frequency are proposed in order to make full use of the measurement information and improve the prediction accuracy. First, by "moving" part of the downlink channel delay to the upload channel, the prediction display problem is transformed into a standard filtering problem and a prediction problem, and the series-connected structure of "observer predictor" is used to solve these two problems. In the observer, the measurement information with multiple delays is reorganized into the form without delay by reorganizing the innovation analysis method, and the measurement information with different frequencies and missing information is processed by the label function. Simulation results show the effectiveness of the proposed predictive display algorithm.
【作者单位】：清华大学自动化系导航所;清华大学深圳研究生院;
【基金】：国家自然科学基金(61673239) 中国博后科学基金(2016T90102) 深圳市科技计划基础研究项目(JCYJ20160428182227081)
【分类号】：TP242

【相似文献】