Design, Control, and Modeling for Soft Growing Robot Deployment for Nuclear Material Inspection

Year
2024
Author(s)
Yimeng Qin - Stanford University
Allison M. Okamura - Stanford University
Marco Salathe - Lawrence Berkeley National Laboratory
Abstract

We present the design of a soft continuum robot that grows via pressure-driven eversion, referred to as a “vine robot”, toward deployment in radiation-contaminated environments. Vine robots mimic plant-like tip extension by emitting material that turns inside out via internal fluid pressure. With a compliant body constructed from durable fabric and unique tip growth mechanism, vine robots can be deployed into narrow, cluttered, and tortuous spaces due to propulsion transmitted at the tip and minimal friction with respect to the environment. Unlike traditional robots that require onboard power and communication, the vine robot centralizes operation and deployment at its base, which is located away from the robot tip. The base station compactly stores the fabric material, deploys it in a controlled fashion, and includes mechanisms to enable steering of the robot body. Sensing systems integrated at the robot tip enable real-time inspection of unknown environments, data collection, and robot localization and mapping. The vine robot system can be operated via human-in-the-loop teleoperation or autonomous control. We designed specialized vine robots with three-dimensional growth and steering capabilities and used them to acquire remote video data inside pipes and around and over mock nuclear material containers.