基于协作机器人仿真环境的关节轨迹预测方法
DOI:
作者:
作者单位:

1.中国科学院沈阳自动化研究所;2.沈阳新松机器人自动化股份有限公司

作者简介:

通讯作者:

中图分类号:

TP242.6

基金项目:

国家重点研发计划


Joint Trajectory Prediction Method Based on Collaborative Robot Simulation Environment
Author:
Affiliation:

1.Shenyang Institute of Automation, Chinese Academy of Sciences;2.Shenyang SIASUN Robot & Automation Co., LTD

Fund Project:

National Key R&D Program of China

  • 摘要
  • |
  • 图/表
  • |
  • 访问统计
  • |
  • 参考文献
  • |
  • 相似文献
  • |
  • 引证文献
  • |
  • 资源附件
  • |
  • 文章评论
    摘要:

    安全在人机协作过程中是至关重要的,必须实时掌握人的行为信息,并进行准确高效的预测。本文基于Linux和ROS系统搭建仿真环境,通过Xtion PRO LIVE深度相机采集多组人体关节的空间位置信息,然后通过无监督学习方法对采集到的坐标点进行聚类和预测,实时更新预测模型,并基于minimum-jerk对特殊异常轨迹进行预测。为了充分保证人的安全,主要研究手部和肘部运动轨迹的预测方法。最终实验结果证明,本文所提出的分层轨迹预测框架可以很好的描述人体运动轨迹,并实时做出准确的预测,不仅保证了人体安全,而且对于提高生产效率具有重要意义。

    Abstract:

    Security is of vital importance in human robot collaboration, it is necessary to realize human behavior in real time and make accurate and efficient prediction. In the paper, a simulation environment was built based on Linux and ROS system. The spatial position of multiple groups of human joints was collected by depth camera of Xtion PRO LIVE. Then, the unsupervised method was used to cluster and predict the collected coordinate points, and the prediction model was updated in real time, minimum-jerk was used to predict the special abnormal trajectory. To ensure the safety of human, the prediction methods of hand and elbow movement trajectory were studied. The final experimental results show that the hierarchical trajectory prediction framework proposed in the paper can describe the trajectory of human better and make accurate prediction in real time, which not only ensures the safety of human, but also plays an important role in improving production efficiency.

    参考文献
    相似文献
    引证文献
引用本文

康杰,贾凯,邹风山,等. 基于协作机器人仿真环境的关节轨迹预测方法[J]. 科学技术与工程, 2019, 19(31): 180-184.
kangjie,,zoufengshan, et al. Joint Trajectory Prediction Method Based on Collaborative Robot Simulation Environment[J]. Science Technology and Engineering,2019,19(31):180-184.

复制
分享
文章指标
  • 点击次数:
  • 下载次数:
  • HTML阅读次数:
  • 引用次数:
历史
  • 收稿日期:2019-01-11
  • 最后修改日期:2019-07-22
  • 录用日期:2019-04-26
  • 在线发布日期: 2019-11-20
  • 出版日期:
×
律回春渐,新元肇启|《科学技术与工程》编辑部恭祝新岁!
亟待确认版面费归属稿件,敬请作者关注