BRC FY17 Topic 5 Octopus-Inspired Autonomous Arms for Soft Robots with Adaptive Motions
Abstract
Octopus Neuromuscular-Inspired Autonomous Arms for Soft Robots with adaptive MotionsWe propose to borrow the octopus~ wisdom used in its neuromuscular distributed sensing and actuation to design, rapid prototype and control robust, energy-efficient, autonomous soft arms. The soft arms composed of force-sensitive hydrogel ~muscles~ with an embedded silver ~neuron~ network can adaptively interact with local environment by local sensing and continuous deformation. The dynamic control via built-in local sensing-actuation feedback loop enablesadaptive reconfiguration to perform high-level tasks - grasping, twisting and propulsion without central control.We propose to create a framework for rapid prototyping and control of robust, energyefficient, autonomous soft tentacles with distributed neuromuscular-inspired sensing and actuation. The tentacles will be capable of continuous deformation through the use of hydrogel~muscles~ and distributed sensing through the use of embedded silver ~neuron~ interconnections. Such a unique octopus-inspired design avoids multiplexed electronic units and high energy input. In addition, the proposed design forms a built-in local ~sensing-actuation~ feedback loop to achieve adaptive reconfiguration and perform high-level tasks such as locomotion and reversible adhesion without direct control from a central nervous system. Our interdisciplinary team consists of a biologist with specialization in functional anatomy (Fisher), a material scientist with specialization in stimuli-responsive polymers and multi-material 3D printing (He), a bio-inspired roboticist (Marvi), a specialist in design and rapid prototyping (Aukes), a specialist in the dynamics and scalable control of bio-inspired robotic swarms (Berman), and a specialist in control of PDEs and infinite-dimensional systems (Peet). To achieve the proposed outcome, our team will execute the following proposed tasks: Biology (Aim 1) We will characterize theneuromuscular structure of octopus tentacles and investigate the roles of distributed sensing, dynamics, and control in their adaptive control-based motions; Material Science (Aim 2) We will fabricate hybrid polymer-based flexible structures using stimuli-responsive hydrogel ~muscles~ with embedded metallic ~neurons~ via a novel additive-manufacturing,stereolithographic 3D-printing technique, which can print both polymeric and metallic materials into highly complex architectures at resolution of up to 3 ~m; Distributed Sensing, Dynamics, and Control (Aims 3-4) We will develop a PDE-based control-theoretic representation of tentacle motion and environmental interaction which is inherently infinite-dimensional and isbased on anatomical studies of octopus tentacles. We will design scalable, consensus-based localfeedback strategies for autonomous performance of high-level tasks such as locomotion in unstructured environments and optimal switchable adhesion. Design Integration (Aim 5) We will design a robust framework for integration of multiple tentacles with a central mantle for high-level communication of commands. We will conduct computational simulations to optimize the system integration. We will also investigate the ways in which material and geometric design considerations of this hybrid material influence the actuation and control capabilities of a design. This work will provide the foundations for soft robotic applications that require safe operation close to humans and resilience to contact with obstacles or falls, as well as improved, compliantsensing surfaces for grasping, manipulation, propulsion, and prosthetics applications.
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- Jan 04, 2017
- Source ID
- N000141712117
Entities
People
- Ximin He
Organizations
- Office of Naval Research
- United States Navy
- University of California, Los Angeles