智能科学与技术学报 ›› 2019, Vol. 1 ›› Issue (3): 280-286.doi: 10.11959/j.issn.2096-6652.201933

• 学术论文 • 上一篇    下一篇

人在回路的混合增强智能在Sawyer的研究与验证

付海军1,2,陈世超1,3,林懿伦1,熊刚1,胡斌1()   

  1. 1 中国科学院自动化研究所复杂系统管理与控制国家重点实验室,北京 100190
    2 中国科学院大学人工智能学院,北京 100049
    3 澳门科技大学,澳门 999078
  • 修回日期:2019-08-01 出版日期:2019-09-20 发布日期:2019-12-17
  • 作者简介:付海军(1993- ),男,河南信阳人,中国科学院大学人工智能学院硕士生,主要研究方向为边缘计算。|陈世超(1987- ),男,山东日照人,中国科学院自动化研究所复杂系统管理与控制国家重点实验室助理研究员,主要研究方向为平行感知、工业物联网、机器学习、智能制造等。|林懿伦(1989- ),男,广东广州人,中国科学院自动化研究所助理研究员,主要研究方向为机器学习、智能机器人、智能交通系统等。|熊刚(1969- ),男,四川乐山人,中国科学院自动化研究所复杂系统管理与控制国家重点实验室研究员,主要研究方向为复杂系统平行控制、智能制造、智能交通。|胡斌(1976- ),男,山东烟台人,中国科学院自动化研究所助理研究员,主要研究方向为智能机器人、无人机、社会制造等。
  • 基金资助:
    国家重点研发计划基金资助项目(2018YFB1702701);国家自然科学基金资助项目(61773381);国家自然科学基金资助项目(61773382);国家自然科学基金资助项目(61533019);国家自然科学基金资助项目(61872365);北京市自然科学基金资助项目(4182065)

Research and validation of human-in-the-loop hybrid-augmented intelligence in Sawyer

Haijun FU1,2,Shichao CHEN1,3,Yilun LIN1,Gang XIONG1,Bin HU1()   

  1. 1 The State Key Laboratory of Management and Control for Complex Systems,Institute of Automation,Chinese Academy of Sciences,Beijing 100190,China
    2 School of Artificial Intelligence,University of Chinese Academy of Sciences,Beijing 100049,China
    3 Macau University of Science and Technology,Macau 999078,China
  • Revised:2019-08-01 Online:2019-09-20 Published:2019-12-17
  • Supported by:
    The National Key Research and Development Program of China(2018YFB1702701);The National Natural Science Foundation of China(61773381);The National Natural Science Foundation of China(61773382);The National Natural Science Foundation of China(61533019);The National Natural Science Foundation of China(61872365);Beijing Natural Science Foundation(4182065)

摘要:

机器学习根据历史数据的模式预测未来,近年来进行了大量研究并得到了应用,但在处理动态、非完整、非结构化信息上与人类相去甚远。为此,引入人的决策,结合机器学习、知识库,构建了一个人在回路的混合增强智能闭环系统。基于 Sawyer 协作机器人搭建了人机融合实验平台,设计了机器人抓取实验。实验结果表明,相比单一机器学习方式,在引入人类智能后,Sawyer在应对非结构化环境下的抓取任务中表现更佳。

关键词: 人机融合, Sawyer, 协作机器人

Abstract:

Machine learning predicts the future through the patterns of past data,and has gained a lot of research and application in recent years.However,it’s far away from human in dynamic,non-complete,and unstructured information processes.Therefore human decision-making,combined with machine learning,knowledge base were introduced in this paper and a human-in-the-loop hybrid-augmented intelligence closed-loop system was built.Based on Sawyer collaborative robot,a human-machine collaboration experiment platform was built,and a grasping experiment was designed.It turns out that the Sawyer,which is introduced human intelligence,performs better in dealing with unstructured environments than that only machine learning is used.

Key words: human-machine collaboration, Sawyer, collaborative robot

中图分类号: 

No Suggested Reading articles found!