Collaborative Research: Embedded Mechano-Intelligence for Soft Robotics

This Foundational Research in Robotics (FRR) award supports a collaborative research project that will create soft materials with integrated sensors, interconnections, logic circuits, and actuators. These materials will enable the development of soft robots with perception, processing, and responsiveness distributed throughout their structures. These capabilities will be demonstrated by a soft robotic platform that can recognize and sort simple objects by their shape, size, and weight, using only these novel materials and without requiring any additional sensing or computing. This soft material-based manipulator will be the first step towards a new class of soft robotic components that can coordinate and intelligently engage with objects in their environment, for a range of practical purposes. Such devices represent a major step for the field of soft robotics and will broadly advance the future of motion control systems, autonomous haptic devices, self-aware sensor-actuator networks, and more. Potential impacts may be felt in societally important application areas such as manufacturing, transportation, and biomedical devices. The research will be coupled with an extensive outreach, education, and mentoring program that integrates the research concepts into classroom and engagement activities among multiple diverse student groups.The research goal of this project is to establish a fundamental synthesis of material and functional components in soft matter to embody intelligence, endowing robots with new capabilities that will significantly enhance their autonomy as compared to the current systems that heavily depend on add-on hardware. The new system will require less electric power and have faster reactions and better survivability than current systems. This project will culminate in a soft robotic sorting manipulator that autonomously detects physical characteristics of items and positions those items into proximity with objects having similar features. This goal will be achieved by a novel integration of embedded mechano-intelligence and field-responsive polymers. Together, these constituents will process information regarding item shape and weight and will trigger reconfiguration of the manipulator so as to position the items into distinct categories. By requiring only a low-voltage input to function, the embedded mechano-intelligence employs only the necessary computational power and eliminates conventional controllers and failure-prone electrical wiring in soft materials. The field-responsive polymers will be used in conjunction with principles of elastic stability theory to minimize the actuating authority required to reconfigure the load-bearing manipulator.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.