MRS Meetings and Events

 

EL05.06.08 2023 MRS Fall Meeting

Soft Waveguide Segments for Deformation Sensing

When and Where

Nov 29, 2023
11:00am - 11:15am

Hynes, Level 2, Room 203

Presenter

Co-Author(s)

Cindy Harnett2,Tram Nguyen1,Michael Han2

Hollins University1,University of Louisville2

Abstract

Cindy Harnett2,Tram Nguyen1,Michael Han2

Hollins University1,University of Louisville2
This presentation looks at soft optical waveguides made from oils and gels for their potential in measuring curvature along paths that bend and twist. Mapping curvature along a path helps capture the shape of soft deformable objects. The focus of this presentation is on creating a defective segment in an otherwise transmissive optical waveguide, a segment with a light transmission level that is highly sensitive to deformation. Optical methods like this one are increasingly popular in soft robotics because their all-polymer construction matches the mechanical characteristics of the surface, and optical signals are less susceptible than electrical signals to temperature drift and electromagnetic interference. Aside from a few recent examples of distributed optical sensors, most examples are point sensors with one highly sensitive section, or integrating sensors that report deformation at an unknown location along the waveguide.<br/><br/>In the first implementation to be discussed, we change a point sensor into a deformation mapper by moving a sensitive segment along the waveguide. Here, the deformation-sensitive section is an air bubble in an oil-filled tube routed along a surface. Loss of total internal reflection at the air bubble means that when the bubble is positioned at a sharp curve, the overall transmission of the channel drops more than when the bubble is in a straight section. The signal is dominated by events at the sensitive segment, while the rest of the waveguide serves as a high transmission signal delivery channel. By circulating the oil through the waveguide with a pump, the sensitive segment is scanned over the waveguide to map out its local curvature, whether that curvature is caused by the static waveguide path layout or by dynamic bending motions of the underlying surface. We demonstrate that the scanned bubble is able to detect and localize the bending location of a soft robotic limb to within approximately 4 cm.<br/><br/>However, the scanned bubble method has a speed limit determined by the Weber number. Scanning faster than a few cm/s causes the air bubble to break up leading to an unreliable signal. Soft segments of silicone or hydrogel are investigated as an alternative low-index segment that can be scanned around the waveguide faster without breaking up. A related topic included in this presentation is verifying the location of scanned segments with a closed loop method. Measuring volumetric oil displacement gives segment location when things are flowing smoothly, but it is possible for solid segments to get stuck. We therefore investigate light reflection as a method to locate segment edges in the liquid core waveguide.<br/><br/>Twist is another aspect of path shape that isn’t captured by a rotationally symmetric, cylindrical soft waveguide segment. We discuss soft waveguide segments with non-circular cross sections for their ability to detect left- and right-handed twisting along the waveguide path. In this example, the application is monitoring the state of a reconfigurable soft robotic actuator.

Symposium Organizers

Michael Ford, Lawrence Livermore National Laboratory
Cindy Harnett, University of Louisville
Juejun Hu, Massachusetts Institute of Technology
Seungwoo Lee, Korea University

Publishing Alliance

MRS publishes with Springer Nature