巡检机器人人机交互中的操作行为分析  

Operation behavior analysis for human-robot interactionof an inspection robot

作　　者：唐立军 李贞辉[2,3] 袁兴宇 李维 刘爱华 TANG Lijun;LI Zhenhui;YUAN Xingyu;LI Wei;LIU Aihua(Electric Power Research Institute of Yunnan Power Grid Co.,Ltd.,Kunming 650217,China;State Key Laboratory of Robotics,Shenyang Institute of Automation,Chinese Academy of Sciences,Shenyang 110016,China;Advanced Technology Institute,Zhejiang University,Hangzhou 310027,China)

机构地区：[1]云南电网有限责任公司电力科学研究院,昆明650217 [2]中国科学院沈阳自动化研究所机器人学国家重点实验室,沈阳110016 [3]浙江大学先进技术研究所,杭州310027

出　　处：《现代制造工程》2021年第3期28-34,共7页Modern Manufacturing Engineering

摘　　要：为了实现巡检机器人人机间的良好交互,保证操作行为的合理性,提出了一种基于贝叶斯网络的巡检机器人人机交互中的操作行为分析方法。首先,基于贝叶斯网络构建巡检机器人操作意图推理模型,对操作人员的操作意图和操作趋势进行推理,以帮助操作人员决策;其次,根据巡检机器人的操作规则库和操作趋势评估当前操作行为的合理性,避免操作人员的误操作行为。实验表明所提方法实现了巡检机器人操作行为意图的快速推理功能,对操作行为的合理性能够进行有效评估。Aiming at the operation behavior rationality in the human-robot interaction of a power line inspection robot,an operation behavior analysis method based on Bayesian networks was presented.Firstly,a model for reasoning the intention of the operation behavior was established on Bayesian networks,and the operation behavior intention and trend were reasoned to help the operator to make decisions.Secondly,the operation behavior rationality was evaluated to aviod the misoperation with the operation rules and operation trends of the robot.The results of experiments show that the operation behavior intention can be reasoned quickly and the operation behavior rationality can be established availably by the method mentioned above.

关 键 词：巡检机器人 贝叶斯网络 操作行为分析 人机交互 

分 类 号：TP24[自动化与计算机技术—检测技术与自动化装置]