Abstract
Creating soft robots with sophisticated, autonomous capabilities requires these systems to possess reliable, on-line proprioception of 3D configuration through integrated soft sensors. We present a framework for predicting a soft robot's 3D configuration via deep learning using feedback from a soft, proprioceptive sensor skin. Our framework introduces a kirigami-enabled strategy for rapidly sensorizing soft robots using off-the-shelf materials, a general kinematic description for soft robot geometry, and an investigation of neural network designs for predicting soft robot configuration. Even with hysteretic, non-monotonic feedback from the piezoresistive sensors, recurrent neural networks show potential for predicting our new kinematic parameters and, thus, the robot's configuration. One trained neural network closely predicts steady-state configuration during operation, though complete dynamic behavior is not fully captured. We validate our methods on a trunk-like arm with 12 discrete actuators and 12 proprioceptive sensors. As an essential advance in soft robotic perception, we anticipate our framework will open new avenues towards closed loop control in soft robotics.
Highlights
S OFT robotics represents an auspicious new paradigm for designing robots with improved adaptability, resilience, safety, and more by introducing compliance and deformability in robot bodies [1]
In [6], a soft robotic hand can predict flexion, lateral, and twist deformations in its fingers, and in [17], a recurrent neural network (RNN) is used to learn the tip position and contact forces of a single soft actuator. Neither of these works achieve a description of full 3D configuration, which is essential for a complete understanding of soft robot behavior and, its eventual control. To address these interwoven challenges, we present a framework for rapidly equipping existing soft robots with distributed, soft, piezoresistive sensors, and enabling them to perceive their 3D configuration via deep learning
We have developed a framework for learning 3D configuration in a soft robot through distributed proprioception enabled by a soft sensor skin
Summary
S OFT robotics represents an auspicious new paradigm for designing robots with improved adaptability, resilience, safety, and more by introducing compliance and deformability in robot bodies [1]. It is this collection of enabling material properties that complicates approaches to soft robotic control. Date of publication February 26, 2020; date of current version March 9, 2020. Truby and Cosimo Della Santina contributed to this work.)
Sign-in/Register to view highlights & summary 
Published Version (
Free)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call