Callaghan Innovation Research Papers

Back to Research Papers

TitleMulti-functional dielectric elastomer artificial muscles for soft and smart machines
Publication TypeJournal Article
Year of Publication2012
AuthorsAnderson, I.A., Gisby T.A., McKay T.G., O'Brien B.M., and Calius E.P.
JournalJournal of Applied Physics
Date Published2012
ISSN00218979 (ISSN)
KeywordsActuators, Actuators and sensors, Artificial muscle, Building blockes, Closed-loop control, Complex task, Control challenges, Conventional robots, Data processing, Degrees of freedom (mechanics), Dielectric elastomers, Drive systems, Elastomers, Electrical parameter, Electro-magnetic motors, Fully integrated, High stiffness, Infinite numbers, Low density, Mechanics, Memory element, Multi-functional, Muscle, Number of degrees of freedom, Piezo-resistive, Plastics, Precise control, Rotational degrees of freedom, Sensory data, Sensory feedback, Shellfish, Smart actuation, Smart machines, Soft actuators, Soft machine, Stiffness
AbstractDielectric elastomer (DE) actuators are popularly referred to as artificial muscles because their impressive actuation strain and speed, low density, compliant nature, and silent operation capture many of the desirable physical properties of muscle. Unlike conventional robots and machines, whose mechanisms and drive systems rapidly become very complex as the number of degrees of freedom increases, groups of DE artificial muscles have the potential to generate rich motions combining many translational and rotational degrees of freedom. These artificial muscle systems can mimic the agonist-antagonist approach found in nature, so that active expansion of one artificial muscle is taken up by passive contraction in the other. They can also vary their stiffness. In addition, they have the ability to produce electricity from movement. But departing from the high stiffness paradigm of electromagnetic motors and gearboxes leads to new control challenges, and for soft machines to be truly dexterous like their biological analogues, they need precise control. Humans control their limbs using sensory feedback from strain sensitive cells embedded in muscle. In DE actuators, deformation is inextricably linked to changes in electrical parameters that include capacitance and resistance, so the state of strain can be inferred by sensing these changes, enabling the closed loop control that is critical for a soft machine. But the increased information processing required for a soft machine can impose a substantial burden on a central controller. The natural solution is to distribute control within the mechanism itself. The octopus arm is an example of a soft actuator with a virtually infinite number of degrees of freedom (DOF). The arm utilizes neural ganglia to process sensory data at the local arm level and perform complex tasks. Recent advances in soft electronics such as the piezoresistive dielectric elastomer switch (DES) have the potential to be fully integrated with actuators and sensors. With the DE switch, we can produce logic gates, oscillators, and a memory element, the building blocks for a soft computer, thus bringing us closer to emulating smart living structures like the octopus arm. The goal of future research is to develop fully soft machines that exploit smart actuation networks to gain capabilities formerly reserved to nature, and open new vistas in mechanical engineering. © 2012 American Institute of Physics.

Back to top