Loading…

Topology optimization applied to the design of actuators driven by pressure loads

Soft robots are intrinsic compliant machines capable of realizing complex movements by the deformation of their own body. They promise to overcome limitations presented by rigid robots in unstructured environments and in handling fragile objects. One of the recent interests in soft robotics is the d...

Full description

Saved in:
Bibliographic Details
Published in:Structural and multidisciplinary optimization 2020-05, Vol.61 (5), p.1763-1786
Main Authors: de Souza, Eduardo Moscatelli, Silva, EmĂ­lio Carlos Nelli
Format: Article
Language:English
Subjects:
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Soft robots are intrinsic compliant machines capable of realizing complex movements by the deformation of their own body. They promise to overcome limitations presented by rigid robots in unstructured environments and in handling fragile objects. One of the recent interests in soft robotics is the development of techniques to design improved soft actuators. In this context, this work presents a density-based topology optimization formulation to design soft actuators driven by pressure loads. This formulation solves design-dependent load problems using mixed displacement-pressure finite elements to synthesize compliant mechanisms actuated by pressure loads. The objective of the optimization is to maximize the output displacement in a given direction while constraining the material volume. A projection technique is proposed to circumvent the appearance of open designs which are of non-interest. The optimization is solved by using an interior point algorithm (IPOPT). Three numerical examples are explored to evaluate the effectiveness of the formulation: a bending, a linear, and an inverter actuator. The numerical results show that the proposed projection technique is successful in finding optimized closed designs for the studied problems.
ISSN:1615-147X
1615-1488
DOI:10.1007/s00158-019-02421-5