Vibrobots (and bristlebots) are simple robots that use a tiny pager / cellphone vibrator motor (with an eccentric weight) to randomly bounce around -- they are the subject of many Maker / DIY projects as well as some well-known commercial toys (such as the $7.00 Hexbug Nano).  Naghi Sotoudeh, a Hizook reader from Iran, contacted us about his latest project: a remote controlled micro-scale vibrobot (measuring just 18 x 12 x 10 mm) that uses two vibrator motors to achieve steerable motion without any wheels.  Naghi's design is similar to some previous steerable vibrobots (eg. the Harvard Kilobot project), but the vibrator motor arrangement gives his design a nice, distinctive faux-wheel look.  The hardware is fairly simple: a small PCB, two vibrator motors, a microcontroller, an infrared photodiode, and a very small battery.  In general, the software for this type of robot isn't too bad either.  In short, this could be a great DIY project and potentially a nice mass-market product.  What do you think...  Would you fork over your hard-earned money for a RC vibrobot kit or pre-built RC vibrobot? 

 

Naghi's Steerable, Remote Controlled (RC) Vibrobot

 

Updated 10/16/2013: Naghi et.al. have improved their vibrobot and released a journal paper about it: "Design and motion analysis of vibration-driven small robot Rizeh."  You can read more about the new robot on their website.

Tiny steerable bristlebot

 

Original: Here are some photos and a video of the Naghi's prototype RC vibrobot.

Remote Controlled RC Vibrobot Remote Controlled RC Vibrobot

On his project homepage, Naghi has a schematic that illustrates the simplicity of the design (shown below).  By controlling the motor velocities using pulse-width modulation, you can make the robot more forward and/or turn.  Presumably, reversing the motor polarity (eg. by tying the grounded motor lead to unused pins on a higher pincount microcontroller instead) we could reverse the direction of the motor rotation and get backwards motion (and turning) too. 

Remote Controlled RC Vibrobot Schematic

 

The robot is remote controlled via infrared (IR) signals received by the IR photodiode.  This solution is low cost, easy to implement, and most people have readily available IR transmitters just lying around (your TV remote).  I'm not sure which IR encoding Naghi's prototype is using, but I know that adding infrared remote control is a fairly straight forward proposition.  Back in undergrad (goodness, that was seriously 6-7 years ago!?), I built a small robot that was controlled by a Sony TV remote.  My implementation was on a PIC microcontroller in assembly (yikes!), and I implemented it all in only about 8 hours (the night before the semester project was due) -- so it's all very approachable.

 

 

Related Work on Steerable Vibrobots

 

Steerable vibrobots (and bristlebots) aren't exactly new.  There have been several different designs that are steerable and/or remote controlled.  From academia, the Kilobots and I-Swarm robot swarm projects use large groups of steerable vibrobots -- videos embedded below.

 

 

 

Other examples come from hobbyists, Makers, and DIYers.  For example, some use a single vibrator motor and weight redistribution to achieve steering.   Others, like Naghi's design, use multiple motors for light seeking (left from 2006) or just RC control (middle from 2010, and right from 2008):

BEAM dual-motor vibrobot  Steerable RC Bristlebot Steerable RC Bristlebot

The one on the right is my favorite.  It is pretty much a standard Bristlebot, but it has the motors offset from the center axis.  It kind of reminds me of a dual-outboard motorboat.  Here's video of it in operation.

 

So anyway, Naghi's idea isn't really new, but the pancake-type pager motors give it a nice, distinctive faux-wheel look.  Like all the others, Naghi's design is incredibly small and compact, has a very low parts cost, and could be a good platform to learn the ins and outs of small robot design for a DIYer (eg. basic circuits and embedded programming).

 

 

Vibrobots, Bristlebots, and the Hexbug Nano -- A Little History

 

For the unfamiliar, vibrobots are an old concept.  I'm not sure how far back the designs go, but I built my first BEAM vibrobot back in 6th or 7th grade (1995ish).  In fact, this was probably one of the first (ever) robots that I built, and it's what got me started in robotics and electronics.  It was a basic 1381-type solar engine design.  The solar cell would store up charge in the capacitors until a preset voltage was reached (detected by the 1381), and it would then completely discharge all the stored energy through the motor, creating a long (30 second) bucking bronco.  I still have that robot (pictured below).  I just tested it out by putting it in the sun, and it still works.  The cats came up to check it out while charging.  When it discharged, it scared the bejesus out 'em.  Ha!

Travis' Early BEAM Vibrobot

Anyway, in 2007 by Evil Mad Scientist Laboratories (EMSL) built a minimalistic vibrobot that was just a pager motor connected to a coincell battery, mounted atop a toothbrush head.  They coined it the "Bristlebot" -- shown below.  So essentially, Bristlebots are a subset of of Vibrobots.  In 2009, there was some controversy when Klutz published a book "Invasion of the Bristlebots," packaged with an almost exact replica of the EMSL Bristlebot (even using the name!) without first consulting the EMSL folks.

BristleBot by Evil Mad Scientist Labs

 

 

Also in 2009, Innovation First released a commercial version of the Bristlebot called Hexbug Nano (shown below).  [ Sidebar: Do any of you guys know if there is a financial relationship between EMSL and Innovation First?  How about the details? ]   Anyway, it was named one of the top 10 holiday toys of the year in 2009 by Forbes.  Right now, Hexbug Nanos retail for as little as $7.00 each.  There are also plenty of accessories available, and these products are commonplace in most toy stores.

Hexbug Nano Bristlebot  Hexbug Nano Bristlebot  Hexbug Nano Accessory Habitat

 

There are also a fair number of Hexbug Nano hacks.  My favorite is IHeartRobotics jousting Lego ninjas riding atop their Hexbug Nano steeds. Creative.

Hexbug Nano Jousting Lego Ninjas  Hexbug Nano Jousting Lego Ninjas

 

Naturally, with such a simple design and meteoric success, copycats will come out of the woodwork.  In 2010 a knockoff appeared, prompting Innovation First to file an intellectual property lawsuit against their new competitor.  Lem at RobotDreams wrote a good breakdown of the situation.   I really don't know what to make of that... The Hexbug Nano concept has been around for so long that I cannot imagine Innovation First snagging a utility patent; perhaps they just had a design patent on "cutesy bug vibrobot"?  Then again, IANAL.

Hexbug Nano Knockoff

 

 

 

The First Hizook Product?

 

Like I said, a steerable RC vibrobot could be a great DIY kit and potentially even a nice mass-market product.   What do you think...  Would you buy a RC vibrobot kit or pre-built RC vibrobot?  I could easily see a kit retailing for under $15, and a full pre-built robot for under $20.  Who knows, maybe something like this could one day become a real product.

 

To express  interest in this type of product (if you could imagine forking over your hard-earned money to buy one)  OR if you're just interested in these types of cool robot projects, toys, or news stories...  please let us know in the comments.

 

Comments (9)

I Heart Robotics September 09, 2011 at 12:12 AM

I thought IFI put in quite a bit of effort doing design for manufacturing and unlike Klutz it was not simply a complete copy and paste job of EMSL's work. However, I'm really disappointed by the patent lawsuit, it seems a bit heavy handed especially for a company that has benefited from other peoples work in the field. The concept of using elastic beam deflection as a means of locomotion obviously has prior art from EMSL's work. So what else could be patentable? From the photos it looks like the legs on the hexbug and the jitterbug both seem to have the same shape, spacing and are both probably injection molded from silicone. If IFI determined the curvature and spacing from trial and error, which is what their website seemed to indicate, then it seems difficult to prove that it was non obvious. Both companies are buying pager motors, possibly even the same model, in bulk. These spin at about the same speed and have about the same size and mass. If you run a numerical simulation of the dynamic system you will probably find that the optimal leg spacing and shape is a function of the motor and mass distribution. Therefore this seems like trying to patent the length of a tuning fork when someone else did the work of figuring out resonance. IANAL. On the original topic. I like the little steerable robot, it's cute, I wonder if you could fit a buzzer or speaker of some sort on it. Also, I think the Tomy Robo-Q was able to use the same IR sensor/led for communication and for obstacle avoidance. Though, I think you might need one extra part/io for an IR led on the robot to do obstacle detection. http://www.botjunkie.com/2008/10/13/tomy-robo-q-is-34-centimeters-of-bipedaliciousness/

Travis Deyle September 09, 2011 at 03:31 AM

@IHR:  Yeah, IFI did a great job with their design to make a truly mass-market product (there's a lot of work involved to build up those distribution channels!); I especially like their packaging solutions.  But really, my question was: does IFI pay EMSL a small (courtesy) royalty or anything? 

Using the same IR receiver for both comms and obstacle detection (with a IR-LED transmitter) is a nice way to repurpose some of the hardware.  That's nice, but you can get even more clever.  LEDs can act as a light sensor -- in addition to transmitting light, the PN junction of an LED is actually photo-sensitive (a photodiode).  So you can use a single LED to alternate between transmission and reception (half-duplex).  This has been used for user interfaces as a basic proximity touch sensor -- I'm sure it could be used for obstacle avoidance too.  Thus, you could just have a single IR-LED to accomplish everything.

yosh September 27, 2011 at 02:38 AM
cool project, can someone offer a source for such tiny 3.6 NIMH battery?
naghi sotoudeh October 04, 2011 at 02:07 AM
I use battery of mobile phone bluetooth headset,
It's under PCB of robot.
Travis Deyle October 19, 2011 at 03:04 PM

Here's another fun HexBug Nano hack that adds a phototransistor on-off switch:

 

If I had kids, I think HexBug hacking would be a fairly common family activity.

Travis Deyle October 23, 2011 at 11:03 AM

I just happened across another RC BristleBot from 2008 called SocBot. This one is free-form soldered on top of a 16-pin DIP wire-wrap socket and fits on top of a quarter.  Personaly, I like the non-painted version better...

RC Bristlebot  RC BristleBot

 

 

Travis Deyle December 19, 2011 at 01:18 AM

I had a fascinating email conversation with a Dr. Panagiotis Vartholomeos, a progenitor of steerable vibrobot motion characterization.  With his permission, I'm posting (relevant) excerpts of our exchange here.

Panagiotis:

Check out these links:

http://csl-ep.mech.ntua.gr/projects/microrobotic/microrobotic.htm
http://nereus.mech.ntua.gr/Publications.html 

They demonstrate the work done on steerable vibrobots  by a robotics team at the National Technical University of Athens, Mechanical Engineering Dept. According to the  literature they are the first to propose, analyze and develop the motion principle based on vibration motors and to demonstrate an autonomous robot (2005, 2006, 2007). The links include videos of the autonomous robot prototype as well as papers describing the background theory.

 

Travis:

Huh, I didn't know about that work -- very interesting.   I'm a bit skeptical that those were the "first" steerable vibratory robots... I know for a fact that BEAM robotics folks were developing them back in the late 1990's.  Plus, I'm pretty sure that the piezo-vibration techniques (similar to the i-Swarm) also existed pre-2000.

 

Panagiotis:

I am aware of the piezoactuation techniques (stick-slip) that you mentioned.  You are absolutely right, vibration stick-slip mechanisms existed pre-2000.  In fact I have worked for 3 years (2003-2005) in the MiCRoN Project.  This was the predecessor of the I-Swarm (the consortium was almost the same). 

However, I think that according to the literature we were the first to study and develop the centrifugal-force actuators, i.e. to produce stick slip controlled motion using the centrifugal forces of the vibrating motors. We used it to develop autonomous robots for micromanipulation and I have a video of steering a needle in the field of view of a microscope. The bulky electronics of the piezoactuators at that time did not allow for power autonomy of similar size robots (the MiCRoNs for example used a power floor or were tethered).  Most of the work was accomplished in 2004-2007 at the Mech. Eng. Dept of NTUA, under the supervision of Prof. E. Papadopoulos.

Unfortunately, I was not aware of the BEAM team. Their work seems fascinating! But I assume they did not publish it in scientific literature, otherwise I would had found it during my literature search back then.

 

Travis:

Oh wow, I am a big fan of the MiCRoN project.  I cited you guys in one of my first robotics papers: wireless power for robot swarms.  Pleasure to (virtually) meet you.  :-)

You're absolutely correct regarding the BEAM robots -- they may have employed a similar technique, but they were never scientifically evaluated with any rigor (to the best of my knowledge).  The whole scene was (informally) led by Mark Tilden at Los Alamos National Labs; later, many of the philosophies were fed into his toy company, WowWee.  I have a soft spot for BEAM robots... that's how I was self-taught in electronics.

 

Panagiotis: 

Be sure to check out this video of our early work:

 

Travis Deyle July 02, 2012 at 01:55 PM

From BusinessWire... HexBug is taking an interesting approach to competitors in emerging markets: they're teaming up with an Indian company (Anacra Merchandise) to deliver robot toys:

“Copycat products which infringe on our intellectual property are present in a number of emerging markets. One of the most effective ways to address the situation is to make our high quality, safe and incredibly fun products available to children at a compelling price,” said Joseph Astroth, Executive Vice President of Innovation First International. “With an increased demand for authentic toy brands, we are confident that Indian children of all ages will love the HEXBUG brand, and we see Anacra as the best possible local partner to help us get in on the ground floor and serve this important emerging market.”

I still question the legitimacy of their IP... but at least their doing the right thing by competing rather than litigating.

kombizz May 11, 2013 at 02:05 PM

I read Travis' article. It was informative and intersting. I met Mr Sotoudeh in the RoboCup Iran open 2013. His concept of his project was amazing and has lots of future application.

http://www.flickr.com/photos/kombizz/8727712517/

Good luck to all scientists who work for a better constructive future for the mankind