Dexterous Soft Robotic Manipulators Via Spatiotemporally Electroprogrammable Stiffness and Self-Shape Sensing

Abstract

Project SummaryThe objective of this project is to develop advanced soft robotic manipulators with highly distributed actuation andproprioceptive capabilities. Despite extensive studies on various soft robotic actuation technologies, there are two fundamental challenges that need to be resolved to be closer to recreating the levels of dexterity and perception found in nature, such as in octopus arms and elephant trunks. First, the current deformation modes of soft robotic manipulators are highly limited and often prescribed, preventing them from performing complex and flexible tasks. Second, accurate proprioceptive sensing capabilities for complex deformations in unstructured environments are lacking. Our project aims to address these issues by developing soft pneumatic manipulators equipped with an innovative enclosure that allows local strain constraints for distributed actuation, and distributed strain-sensing for proprioception. This enclosure includes a large array of individually controllable millimeter-scale electroadhesive clutches, which can modulate local tensile stiffness through switchable coulombic electroadhesion. Spatiotemporal programming of this distributed electroadhesion will enable dynamic, highly complex modes of deformation, such as bending, stretching, and twisting, when integrated with standard pneumatically driven soft arms. Furthermore, we will develop mesostructure-based highly stretchable capacitivestrain sensors as distributed strain sensing elements, which will be used for high-accuracy 3D reconstruction of complex deformations. Towards this goal, we plan to undertake the following tasks: (I) develop bio-inspired electroadhesive clutch arrays with tunablelocal tensile stiffness; (II) investigate the relationship between strain-limiting patterns and deformation modes of soft manipulators; (III) establish mechanics-driven modeling framework for shape reconstruction of soft manipulators and implement it using distributed, advanced strain sensors capable of accurately measuring local strain. This work, if successful, will significantly improve the capabilities of soft robotic manipulators. This could open up possibilities for highly dexterous manipulators with efficient closed-loop control, suitable for operation in dynamic, unstructured environments.

Document Details

Document Type
DoD Grant Award
Publication Date
Nov 21, 2023
Source ID
N000142412030

Entities

People

  • Hangbo Zhao

Organizations

  • Office of Naval Research
  • United States Navy
  • University of Southern California

Tags

Readers

  • Distributed Systems and Data Platform Development
  • Nanocomposite Materials Science
  • Robotics and Automation.

Technology Areas

  • AI & ML
  • AI & ML - Autonomous Systems
  • Autonomy
  • Autonomy - Autonomous System Control