计算机科学
材料科学
机械工程
声学
工程类
物理
作者
Zehao Huang,LI Liang-cai,Peng Wang,Zhiguang Song,Yongbo Shao
摘要
In this paper, a non-contact multi-degree-of-freedom (multi-DOF) attitude measurement method for spherical joints based on magnetic sensing is proposed, which realizes attitude detection of spherical joints without introducing new contact force and shows advantages of easy installation, low cost, and strong practicability. The sensor system utilizes theoretical derivation, data fitting, and machine learning methods to establish a function mapping model between magnetic field variation and Euler angles of spherical joints. In the research, firstly, the optimal placement scheme of sensors is solved by numerical simulation. Then, data fitting and improved GA-BP neural networks are adopted to get the inverse solution model of spherical joints. Finally, simulation results show that the proposed measurement method is feasible and can be applied to large-scale production.
科研通智能强力驱动
Strongly Powered by AbleSci AI