Electroactive Polymers (EAP) as Artificial Muscles (EPAM) for Robot Applications

Electroactive Polymer (EAP) Artificial Muscle Robot called MERbot from SRI International

Having previously written about various artificial muscle technologies, I'd like to examine the electroactive polymer (EAP) variant in more detail.  I'll briefly discuss how EAPs function, then move on to myriad examples of EAPs used in robotics applications, including: biomimetic robot eyes, childrens' toys, and flapping-wing ornithopters.   I'll also look at electroactive polymer artificial muscles (EPAM) that were invented at SRI International and subsequently spun off to startup Artificial Muscle, Inc.  In my favorite example, a hexapod walker was constructed at SRI whose muscles are used for both structural support in addition to actuation.  Now if they could also function as energy storage devices, they'd be the ultimate biological analog.

Electroactive Polymers (EAP) are a relatively new class of "smart material" that deform in the presence of an applied electric field, much like piezoelectric actuators.  However, unlike piezoelectric actuators, EAPs operate on fundamentally different principals and produce force / strain / deflections more similar to that of biological muscles.  In fact, there is yearly competition sponsored by NASA that seeks an EAP-powered arm that can defeat a human at arm-wrestling!

Wikipedia has a fair article on EAPs.  Of note, there are two types: ionic and and dielectric.  The ionic EAPs operate through the movement of ions within a polymer, as shown in the diagram below.  A webpage by Dr. Bar-Cohen provides numerous links and pointers about ionic EAPs, as does a NASA webpage  that includes recipes to make your own.  There are also several papers about ionic EAP fabrication here and here, though the NASA recipes may be simpler.  For the casual hobbyist, the easiest solution is probably to purchase a kit from Environmental Robots, Inc (ERI)

Electroactive Polymer Concept 

Given my EE background, I understand the dielectric EAPs much better. They are essentially an elastomeric capacitor -- electrostatic forces cause charged electrodes to squish an intermediate polymer layer causing it to expand, as shown in the diagram below. The entire process is also reversible, which can be used to generate electricity or be used as a sensor (much like piezoelectrics).  Dielectric EAPs form the basis of the electroactive polymer artificialmuscle (EPAM) "spring roll" actuators (pictured below-right) developed by SRI International that were ultimately spun off into a startup called Artificial Muscle, Inc.

Electroactive Polymer Concept  Electroactive Polymer (EAP) Artificial Muscle (EPAM) Spring Roll Actuator

A video of the spring roll actuator is shown below.

Six of these spring roll EPAMs were used to build a hexapod called MERbot.  The cool thing about MERbot is that the EPAMs provide both structure and actuation.  From a bio-mimicry standpoint, this is particularly compelling; if the EPAMs could store energy as well, it would be very nearly an "ideal" component.  Check out the MERbot video below.

The EPAMs were also used in a number of other robots at SRI -- check out the videos.


Despite EAPs being in their infancy and rather primitive at this time, I believe that their low-power, compliant actuation makes them well-suited for applications in animatronics.  For example, these biomimetic animatronic eyes from Eamex have extremely realistic motion that is achieved without complex and bulky motor/pulley systems. Be sure to check out the video, embedded below.

In another example, this robotic head shows expressive facial movements using relatively simple actuators.  It was a platform for EAP demonstrations, photographed at the NASA Jet Propulsion Lab (via David Hanson at UT Dallas). Again, be sure to check out the video.

Also from Eamex, toys such as stuffed animals and dinosaurs have been created.  The natural compliance of the EAP actuators adds to their intrinsic safety. Check out the videos (here and here), embedded below.


Despite the infancy of EAP robot designs, researchers are already looking at future applications, such as the EAP-actuated unmanned aerial vehicle in the shape of an ornithopter, or flapping-wing aircraft -- artist's conception below.

Electroactive Polymer (EAP) Ornithopter 

To be designed by the NASA Institute for Advanced Concepts (NIAC), this idea was featured in an IEEE Spectrum article entitled "Fly Like a Bird: Flapping Wings Could Revolutionize Aircraft Design."  In this case, the EAPs are a crucial component of the thin-film laminate stack-up that would contain power harvesting (photovoltaic), power storage (lithium batteries), and actuation / sensing (the EAPs).  

Electroactive Polymer (EAP) Ornithopter

Presumably, such systems could stay aloft for prolonged durations, autonomouslygather scientific data, relay communications,  survey terrain, outmaneuver fixed-wing craft, and/or performing planetary exploration.  This concept isfurther explored in Colozza's slides on Solid State Aircraft.   While this concept is relegated to the future, small EAP flapping-wing craft are certainly being examined.

Electroactive Polymer (EAP) Flapping-Wing Craft    Electroactive Polymer (EAP) Flapping-Wing Craft 


It's my hope that electroactive polymers and associated artificial muscles become more commonplace.  Personally, I'd enthusiastically applaud sheets of dielectric EAPs being made available to hobbyists and researchers for experimentation -- perhaps the folks at Artificial Muscle Inc are listening...?   



I've done quite a bit of research into this material for the biomedical applications, and I'm interested in doing some tests, However I cannot get my hands on any sheets of this material... Is there any way I can contact someone to buy some.



—Werner De Wilzem
indeed, very good question. I'm looking for some to, and can't get my hands on it.....

3m VHB tape (4910) is being used as a dielectric EAP material. The electrodes are made of some sort of binder and graphite or carbon black



—Danny Park
Same question as Werner, Im very interested in experimenting with this material in animatronics, but I can't find any provider or where to buy this, if someone knows and can pass the info would be very apreciated, also I want to know about pricing, how expensive is this material. Thanks

The present generation of artificial muscles do not generate lifting force like a true striated muscle. In fact, they are not  artificial muscles  at all.

In the striated muscle, a myosin myofibril moves up actin myofibrils in multiple incremental steps. This movement is analogous to an inchworm actuator,  in that the total motion is accomplished in short incremental steps, during very quick bursts of activity. The total contractile motion lasts only a few milliseconds.

One result of this approach is that the mysin is capable of lifting a comparatively  larger force  than if it were to  actuate  in one giant  step  (Obviously) ...

Work = Force X Distance

Thus, the myosin may only move a comparatively small amount per increment but pull with large amounts of force;  per unit of supplied energy per increment. Of course, this requires the myosin to move through many cyclical increments to achieve large motion  (at most a few centimeters).

Through multiple cycles of incremental steps,  the myosin achieves both large percentage of overall motion and large values of contractile force.

This  is not  how the present generation of EAP's and many other relatively     weak  so-called artifcial muscles produce movement.


\D.  Gray

—Daniel Gray

On what point are you proving EAP doesn't act as a muscle? That it actuates as a single mass rather then per cell? 

If you research Dielectric elastomers more you will learn they are capable of the same strength and elongation that muscles acheive. This is acheived through a DEA Roll which can have 20 layers or more of EAP. The more rolls you put, the more stress it can acheive. 


Here is a paper talking about the properties of EAP. Figure 2 shows on how Dielectric elastomers(EAP) can acheive muscle actuation. 



Hi, I was wondering how much energy these EAP components/actuators use - i.e would it be possible to run them off a 110ma LiPo battery and if so for how long?

—Kathy Vones

I've received several inquiries about where to purchase or how to build EAPs -- specifically, dielectric elastomer actuator (DEA) EAPs like those in the videos.  

First, let me say: the EAPs in this article (pictures and videos) are not mine.  They were made by SRI or their spinoff company, Artificial Muscle Inc (AMI). If you're an OEM building mass-market products, you should contact AMI.

To the best of my knowledge, no one currently offers kits for experimentation -- neither commercial nor DIY. 

I have never actually built my own DEA EAP, but I've researched the topic extensively and talked to a few professional researchers who have (they provided some helpful hints).  About a 18 months ago, I set out to build a kit for sale right here on Hizook.  To make a long story short, it didn't happen.  I bought all the necessary supplies (muscle materials and power supplies), but then I got completely consumed by other efforts.**  I never had a chance to follow through and finish the development (lab work) to make it happen.

I'm reluctant to share my insights for two reasons, in this order:

(1) There's a serious safety concern with DEA EAPs.  They require high voltages (a few thousand volts) that can potentially be lethal.  This is especially true if you go for the "easy" (cheap, and readily-available) high-voltage supplies!  That's why I was developing my own high-voltage, low-current power supply to drive the DEA EAPs (and hence slightly more costly kit).  I refuse to expose your average DIYer to that kind of (lethal!) risk.  For this reason alone, I'm unwilling to even share the "recipe" (ie. the muscle materials) used to build DEA EAPs.

(2) Plus... I still want to build the kit.  I imagine a kit could retail for somewhere in the $100-$200 range, but don't hold me to that.  Want it to happen?  Help motivate me by leaving a comment or contact me directly via Hizook contact form!  It will really help me gauge interest and (potentially) re-prioritize my time to finish the kit.



** I was busy finishing my PhD.  Then, I got consumed with postdoc efforts (I'm a NSF / CRA "Computing Innovation" Postdoc Fellow at Duke University) and startup efforts (I'm the "lead programming monkey" at my wife's web-based fashion startup).


—Travis Deyle

a kit would certainly be nice.  However, re "They require high voltages (a few thousand volts) that can potentially be lethal", why don't you just use a dirt cheap high voltage module that comes with those air ionization needles?  I assume the casing is grounded and the hot lead is floating in air to ionize it.  You can buy those at any electronics surplus shop for <$15 and just cut off the needles board and you have a 15KV very low current supply that runs off standard 12V DC supply.  You can only get a minor zap from these.  



i am also intertested in EAP and i found this company that can provide some kind of testing packages.

i have not tried them yet but i am thinking to do so soon.




Hi. I am wondering how you evaluate the current state of this field. Are all those problems you are tellling about (high voltage, purchase of material,...) still a issue. Could you please give an update on your work/your view on how things developed over the last years? Thanks


If you need EAP's StretchSense makes capactive stretch sensor kits, they make self powered versions too


soft actuator sensing and soft robotic orthoses using fluidic muscles https://www.youtube.com/watch?v=jHdkw7KYm0s