摘要
随着串联六自由度工业机器人的技术提升,其逐渐被尝试应用于高精度的机械加工场景.机械加工的精度和稳定性受工业机器人的动态特征影响,因此对于机械加工场景中的机器人进行动态特征研究具有重要意义。而工业机器人工作空间大,对整个工作空间进行大量的实验模态测量与分析是一项繁重的工作。因此本文提出一种采用自适应迁移高斯过程进行工业机器人不同工作空间之间实验模态迁移的方法.通过本方法可实现机器人不同工作空间之间实验模态的迁移分析,由足量的源工作空间数据与少量的目标工作空间数据建立准确的目标工作空间实验模态模型:对COMAU NJ220机器人进行实验模态测量与分析,实现源工作空间模态模型向目标工作空间模态模型的迁移,实验结果表明,所采用的实验模态迁移方法既可以减少实验模态测量工作量,又可有效保证目标工作空间实验模态的预测精度。
With the technology development of serial 6-DOF industrial robot,it is gradually applied to high-precision machining.The accuracy and stability of robot machining are affected by its dynamic characteristics,and it is crucial to study the robot dynamic c haracteristics ahead of actual machining.It is a heavy task to measure and analyze its dynamic characteristics in the entire working space.Therefore,this paper proposes an adaptive transfer learning algorithm based on Gaussian processes to transfer experimental modal properties between different workspaces.With the proposed method,the transfer analysis of experimental modal properties between different workspaces can be realized,and an accurate modal properties model of target workspace can be established by sufficient source workspace data and a small amount of target workspace data.The experimental modal of COMAU NJ220 robot is measured and analyzed both in the source workspace and target workspace.The experimental results show that the proposed method can not only reduce the workload of experimental modal measurement,but also effectively ensure the prediction accuracy of experimental modal in the target workspace.
作者
郭皓邦
李钰
叶葱葱
GUO Haobang;LI Yu;YE Congcong(Aecc South Industry Company Limited,Zhuzhou Hunan 412002,China;School of Mechanical Science and Engineering,Huazhong University of Science and Technology,Wuhan 430074,China)
出处
《机械设计与研究》
CSCD
北大核心
2021年第6期13-17,共5页
Machine Design And Research
