Research

Soft Robots at Purdue and Yale University

Aug 2016 - April 2022

In Fall 2016, I joined the Faboratory at Purdue University with Prof. Rebecca Kramer-Bottiglio, to work on soft robotics. We applied compliant materials to robots, toward the goals of increased safety and dexterity. I then moved with the lab to Yale University, where I embedded robotic functionalities into thin skins, and designing shape changing robots. In Spring 2021, I received the Henry Prentiss Becton Prize for my research, and defended my dissertation in December 2021. I then stayed on as a post-doctoral researcher until I moved to Arieca in April 2022, formally receiving my PhD diploma in May 2022. Representative projects include:

1. Making stretchable "jamming skins" that change their stiffness and stretchability upon application of a vacuum (shown below in Figure 1). Very few soft robots can switch their stiffness to interact with the environment; those that can stiffen often require significant design constraints (such as a large volume of stiffness-changing material) which limit their applicability. By embedding strong jamming capabilities into thin sheets, I aimed to allow stiffness-changing capabilities to be added to the full range of soft robots. After I solved the major mechanical and manufacturing challenges, the final jamming skins are now scalable and can be readily integrated into other future soft robots.

   

2. Embedding actuation and sensing into planar "robotic skins" to turn passive deformable bodies - including foam, clothing, and stuffed animals - into robots (shown below in Figure 2). These multi-functional robots could serve as lightweight tools for space missions, while also finding applications as wearable electronics for healthcare here on Earth. Additional coverage can be found on the web, including at Smithsonian.com and NationalGeographic.com.

   

3. Sensors for impact-resistant, lightweight "tensegrity" robots (see the bottom of Figure 2 for an example). Most of my work in this area uses short rods (10-30 cm) connected to stretchable sensors and pneumatic actuators, but we're inspired by our talented collaborators at NASA Ames, including Massimo Vespignani, who has demonstrated a human-sized "Super Ball Bot" that can absorb heavy impacts and even survive drops of several meters. These tensegrities have potential as exploratory robots that can exploit steep terrain changes, rather than navigating around them.
4. Making shape-changing robots with talented collaborators from Yale, the University of Vermont, and Tufts University. Example robots include our cable-driven shape-changing robot and our robot that changes shape to locomote in different environments. Shi En Kim concisely summarized the project on Massive Science.
5. Currently, I'm focused on building stretchable sheets that can sense their shape in 3D. Stay tuned for more information!

Agricultural Robotics at Iowa State

May 2014 - Aug 2016

During my concurrent degrees at Iowa State University (B.S. Mechanical Engineering and M.S. Agricultural Engineering, with thesis) , I worked with Lie Tang on developing robotic systems and image-processing pipelines for data collection during the entire plant life-cycle. This data is useful for improving crop yield and studying the effects of various environmental parameters on plant health.

At first, I assisted with data collection and mechanical design, culminating in my masters' project where I led a small team (a few undergraduates and masters' students) to design and program robots to fulfill the goals of the broader projects. The most sophisticated one (prototype shown below in Figure 3), for the Enviratron project, was a mobile rover with a robot arm and a Kinect V2 3D camera for collision-free probing during use with researchers' specified instruments, such as a fluorometer. We additionally created a slender and compact field robot for 15% of the cost of the commercial alternative, allowing us to collect data on crops, such as corn and soybeans, that are grown in fields with narrow row spacing that conventional field robots could not navigate.