Back to search

Cornell University

PHYSICAL INTELLIGENCE ENABLED BY SOFT ROBOTIC SYSTEMS

Abstract

dc:description.abstract

Soft robotics is of special interest in the study of physical intelligence, aka. the intelligence that predominantly comes from the physical body and its morphology, due to the compliance of the materials and mechanisms. Soft systems have been demonstrated for complex motion, human-like sensing capabilities, morphing, self-healing, adaptation, and growth; which are abilities highly desirable in intelligent agents to endure the demanding working conditions of the real world. This dissertation explores the use of soft robotic systems as platforms to encode intelligent behavior in the body of the robot, and it presents examples of soft robotic system modules that could facilitate perception, controls and actuation in embodied AI agents, thus composing a better embodiment for them. In an advance for robotic haptic perception systems that could allow an agent to understand its embodiment and its interactions with the environment, I present 1) a self sensing membrane that estimate its deformation state based on capacitive readings and supervised learning models; 2) a stretchable optical fiber that maps bending, stretching and location of pressing into intensity and color of light; and 3) an optoelectronic robotic flesh that is scalable to cover the whole body of a robot and encodes deformation, temperature and damage stimuli into light. In an effort to facilitate controls and complex actuation in robots through means of physical intelligence, I present 1) a tendon driven actuator in which the actuation trajectory is programmed by the placement and number of tendons inside its body; and 2) a fluidic hyperelastic textile actuator that can inflate in complex 3D shapes by locally programming the mechanical properties of its membrane using digital embroidery. The challenges of artificial physical intelligence in soft robots, including fabrication, design, and simulation; and demonstrations of the system modules in real world applications are discussed in each chapter.

Degree

thesis:*
Name thesis:degree_name
Ph. D., Systems Engineering
Level thesis:degree_level
Doctor of Philosophy
Discipline thesis:degree_discipline
Systems Engineering
Grantor
Cornell University
Year dc:date.issued
2021

Author and committee

dc:creator, dc:contributor.*
Author dc:creator
  • Barreiros Flores, Jose Antonio
Chair dc:contributor.chair
  • Shepherd, Robert F.
Committee members dc:contributor.committeemember
  • Campbell, Mark
  • Selman, Bart

Subjects

dc:subject × 4

Rights

Language dc:language.iso
en

Identifiers

dc:identifier.*
Dc Identifier Bibid
15160086
Dc Identifier Other
ProQuest Submission ID: 12773
ProQuest Publication ID: 28714065

Chain of custody

source
Harvested from
Cornell University
Base URL
ecommons.cornell.edu/server/oai/request
Last updated
2026-11-19
Source record
OAI-PMH GetRecord
citation

Barreiros Flores, Jose Antonio. PHYSICAL INTELLIGENCE ENABLED BY SOFT ROBOTIC SYSTEMS. Doctor of Philosophy thesis, Cornell University, 2021. https://hdl.handle.net/1813/110505