A Lumped-Mass Model for Large Deformation Continuum Surfaces Actuated by Continuum Robotic Arms
A Lumped-Mass Model for Large Deformation Continuum Surfaces Actuated by Continuum Robotic Arms
复制标题
DOI：
10.1115/1.4045037
复制
发表时间：
2019-10
期刊：
影响因子：
--
通讯作者：
Hossein Habibi;Chenghao Yang;I. Godage;Rongjie Kang;I. Walker;D. Branson
中科院分区：
文献类型：
--
作者：
Hossein Habibi;Chenghao Yang;I. Godage;Rongjie Kang;I. Walker;D. Branson
目前，机器人手臂驱动的柔性表面在医疗保健、汽车、航空航天和制造业等各个领域的机器人应用中受到高度关注和需求。然而，到目前为止，它们的设计和控制主要是基于“试错”方法，需要多次试验和/或高度的用户专业化。因此，实现具有大曲率变形能力的柔性表面的稳健方法需要一个可靠的、经过验证的模型，该模型考虑了许多物理和机械性能，包括弹性、材料特性、重力、外力和厚度剪切效应。这种模型的推导将使基于预测的柔性机器人表面控制方法的进一步发展成为可能。本文建立了连续机械臂驱动下大变形柔性曲面的集总质量模型。所得到的模型包括表面和驱动元件的力学和物理特性，以预测多种曲率方向和驱动配置下的变形。该模型在实验系统中进行了验证，实验和建模结果之间的测量位移显示出相当大的一致性，平均误差约为最终变形形状表面长度的1%。
Currently, flexible surfaces enabled to be actuated by robotic arms are experiencing high interest and demand for robotic applications in various areas such as healthcare, automotive, aerospace, and manufacturing. However, their design and control thus far has largely been based on “trial and error” methods requiring multiple trials and/or high levels of user specialization. Robust methods to realize flexible surfaces with the ability to deform into large curvatures therefore require a reliable, validated model that takes into account many physical and mechanical properties including elasticity, material characteristics, gravity, external forces, and thickness shear effects. The derivation of such a model would then enable the further development of predictive-based control methods for flexible robotic surfaces. This paper presents a lumped-mass model for flexible surfaces undergoing large deformation due to actuation by continuum robotic arms. The resulting model includes mechanical and physical properties for both the surface and actuation elements to predict deformation in multiple curvature directions and actuation configurations. The model is validated against an experimental system where measured displacements between the experimental and modeling results showed considerable agreement with a mean error magnitude of about 1% of the length of the surface at the final deformed shapes.