Great Innovative Ideas are a way to showcase the exciting new research and ideas generated by the computing community.
The following Great Innovative Idea is from Jun Zhang and Michael Yip. Their paper called Three-Dimensional Hysteresis Modeling of Robotic Artificial Muscles with Application to Shape Memory Alloy Actuators was one of the featured talks at the Computing Community Consortium (CCC) sponsored Material Robotics (MaRo) Workshop at the 2017 Robotics Science and Systems (RSS) Conference.
Robotic artificial muscles are actuators that can make robots move. Unlike electric motors, robotic artificial muscles are compliant and can generate straight contractions just like our biological muscles. Recently, they are increasingly popular in many exciting areas, such as biomimetic robots, soft robots, and safe human-robot interaction. To practically use robotic artificial muscles, it is crucial to have an accurate model to understand how they work.
Obtaining an accurate model for different robotic artificial muscles is a challenging task: they often exhibit a type of nonlinearity called hysteresis; the hysteresis is three-dimensional among input, strain, and tension force; and the hysteresis in different artificial muscles is often different. For the first time, we proposed a three-dimensional model that can accurately capture and estimate the performance of different robotic artificial muscles. The proposed model is efficient in computation since it is derived by embedding a two-stage two-dimensional model, rather than employing multiple two-dimensional models conventionally.
This is the first approach to capturing the three-dimensional hysteresis in a class of artificial muscles for robots. The approach is efficient, general and adaptable for different applications: First, the proposed model can be constructed using any existing two-dimensional hysteresis models. Second, the proposed scheme is based on a data-driven model that does not carry physical implications. This study will strongly facilitate modeling and control of muscle-powered robots.
Jun Zhang: My research interests include 1) smart actuators and artificial muscles for mechatronics and robotics including microactuators, microrobots, bio-inspired robots, and assistive robots, and 2) modeling and control of smart actuators and artificial muscles with emphasis on high fidelity and low complexity.
Michael Yip: My research focuses on three areas: (1) flexible robots for surgery, (2) visual computation for image-guided robots, and (3) robotic actuators for bionic devices. Recent efforts involve building, controlling, and automating endoscopic and catheter robots for treating heart and lung disease, designing artificial intelligence for robot-human collaboration in surgery, and augmenting surgeon teams with augmented reality for minimally invasive surgery.