Collaborative Research: Creating System Level Intelligence by Harvesting the Physical Reservoir Computing Power in Origami

Regents Of The University Of Michigan - Ann Arbor, Ann Arbor MI

Investigators

Abstract

This project supports foundational research to create transformative structures and materials with multi-faceted intelligence in the mechanical domain (aka. mechano-intelligence). Such mechano-intelligence will be embodied and synergized throughout the physical body of these structures and materials to execute highly autonomous engineering tasks, such as acquiring information from the surrounding environment, memorizing it, and deciding on an action plan. The novelty of this project is the use of multi-functional origami as a mechanical neural network so that one can harness its computing power as the core foundation for creating and integrating essential intelligent elements. The impact of this project will be the advancement of many intelligent engineering systems widely applicable in different industries, with less power requirements, more direct interaction, and more resilience against harsh environments and cyberattacks. In addition, this project will integrate its research outcomes into new teaching curricula, outreach activities, and lab demonstrations, cultivating diverse students’ interest in STEM pursuits under the inspirational theme of physical intelligence. The vision of this collaborative effort is to create structures and materials with multi-faceted intelligence embodied in the physical body. Although some studies have attempted to distribute (offload) intelligence to the mechanical domain, there is still a lack of a broad and systematic foundation for constructing and integrating the different elements of mechano-intelligence, such as information perception, memorization, and decision-making. To this end, the investigators will bridge this crucial gap by harnessing multi-functional origami as a mechanical neural network and leveraging its physical reservoir computing power as the needed foundation for creating, measuring, and designing the essential elements of intelligence. More specifically, the investigators will 1) explore the extent of complexity and sophistication that mechano-intelligence can attain on physical platforms, 2) formulate performance metrics to quantify and measure mechano-intelligence, 3) correlate the design and mechanical properties of the physical platform to the corresponding intelligence performance, and 4) create a systematic design method for integrating mechano-intelligence with engineering functions. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.