Session: MR-04-01 - Origami-Based Engineering Design
Paper Number: 89501
89501 - Experimental Validation of Origami's Reservoir Computing Power and an Mechano-Intelligent Task of Payload Identification
Animals’ excellent capability to achieve complex locomotion and manipulation behaviors benefits from their unique body morphology, distributed sensory system, and direct connection with the environment, creating intelligence embodied directly in their bodies. This presents a significantly more flexible control strategy than most engineered robots that rely solely on a centralized control unit. A computing frame called physical reservoir computing is put forward to mimic the animal’s embodied intelligence, enabling many mechanical structures and materials to accomplish complex computation tasks. In this paper, we experimentally verify the computing capability of Miura-ori sheets using emulation tasks for the high order nonlinear system and find strategies to improve the computing performance by optimizing their physical and computational designs. Comprehensive parametric studies show that designing an unevenly distributed nodal mass and fine-tuning the origami folding designs are the most effective approach to improve computing performance. We also demonstrate the potential of using the origami reservoir to achieve an intelligent task: predicting the payload magnitude. In other words, the nonlinear dynamics of origami can be utilized to ‘feel’ the weight of objects on top of it. The results of this paper can pave the way to enable more advanced mechano-intelligence in origami-based structures and robots.
Presenting Author: Jun Wang clemson university
Presenting Author Biography: Jun Wang received her master's degree in mechanical engineering from Tongji University in China and became a graduate student and worked as a research assistant in Clemson University from 2020, advised by Dr. Suyi Li. Her research focuses on origami structure with mechano-intelligence.
Authors:
Jun Wang clemson universitySuyi Li Clemson university
Experimental Validation of Origami's Reservoir Computing Power and an Mechano-Intelligent Task of Payload Identification
Paper Type
Technical Paper Publication