Controlling a Robot From a Smartphone's Headphone Jack

RedEaredSlider passes along this excerpt about what looks like a smart advance in controlling hobbyist robots:"The concept is quite simple: put a wheeled chassis on a smart phone or iPod Touch that allows for using the device as the 'brain.' But that simplicity is what makes the robot, called Romo, powerful. Since the controls are contained entirely within the phone, they can be downloaded as apps. One can add new physical capabilities to Romo -– a claw, or a scoop -– but that doesn't require any new additions to the phone. Also, the controls are through the headphone jack. That simplifies the design and means that the robot doesn't need to be linked with only one brand of smart phone."

Re:

[Cryacin]

I can just see all the old Commodore 64 nerds dusting off their old tape recorders and interfacing using audio tones.

Sinclair from futurama might just be just around the corner.

Re:

[hedwards]

I was wondering how this was different than when folks were sending information over a phone line via modem. Other than the obvious fact that they're using a sound jack rather than a modem.

Audio jack to get a standard connector?

[Anonymous Coward]

Why not use micro-USB instead of the audio connector?

Re:

[ewanm89]

Audio is a universal interface, all one needs is a good enough sound card to modulate it (even crappy onboard ones in smart phones are good enough for that these days).

Re:

[Hognoxious]

How does it work? Is there something on the robot that detects pulses of a certain frequency and/or length and interprets them along the lines of "Aha! F#, 33 milliseconds. Stopping."?

I wonder what would happen if you played music into it?

Re:

[ewanm89]

You'll need to do some kind of demodulation and signal processing on the robot end of the link. As for music, it depends on how they are doing their modulation and stuff. if one does it right one can filter off the music, but I expect they aren't even considering that kind of stuff.

Re:

[Hognoxious]

My brother had a toy car that was controlled by a kind of clicker. It would dance to certain records - that was what made me think of music.

Just plug a microphone in & control it with a whistle or bugle calls ;-)

Re:

[hedwards]

Polar has been using a similar system for their heart rate monitors for years. I played around with it, but it tends to be a pain and from what I've observed I think it has issues dealing with the Doppler effect. The only way I could get the data to transmit accurately would by by setting both the mic and the watch down to work.

Re:

[fluffy99]

Polar has been using a similar system for their heart rate monitors for years. I played around with it, but it tends to be a pain and from what I've observed I think it has issues dealing with the Doppler effect. The only way I could get the data to transmit accurately would by by setting both the mic and the watch down to work.

Yeah, had one of those too. The polar SonicLink was very temperamental indeed. You had to have your speakers set just right to send data to the watch, and set the watch in exactly the right position in front of the mic. Their IR interface was much better although the IR dongle they sold was expensive.

Re:

[Anonymous Coward]

One way to do this is via DTMF. There's a very easy to use DTMF decoder chip (MT8870) that can be used for this.
The smartphone generates DTMF, which is fed to a circuit in the microcontroller that decodes it via the MT8870.
The problem would be feedback. The microcontroller won't be able to provide feedback to the smartphone via this method.

Re:

[satuon]

I own a Chinese-made mp3 player that connects to the computer through the stereo jack - it has a cable that has USB at one end and a stereo jack plug at the other. I could even use it like a USB memory stick with this cable. And it's charged through the stereo jack - the adapter ends with a stereo jack plug.

I think the stereo jack is just used as a conductor to carry electric current, both for charging and to carry information. They're doing it as a cost-cutting measure, they save-up on 2 additional ports

Re:

[fluffy99]

I own a Chinese-made mp3 player that connects to the computer through the stereo jack - it has a cable that has USB at one end and a stereo jack plug at the other. I could even use it like a USB memory stick with this cable. And it's charged through the stereo jack - the adapter ends with a stereo jack plug.

I think the stereo jack is just used as a conductor to carry electric current, both for charging and to carry information. They're doing it as a cost-cutting measure, they save-up on 2 additional ports that way. So when I read about this article I wasn't really surprised.

Totally different. This setup is using a sound signal to transmit commands.

The Apple shuffle had a special usb-headphone cable that had an extra ring for the usb. There is also a chip used in some of the newer that doesn't require this extra connection, and senses when it's plugged into USB and switches over. That eliminats the need for an extra usb connection although forces you to use the special cable instead of an off-the-shelf usb cable.

Re:

[Anonymus]

I imagine it works almost exactly like a modem. Don't you remember accidentally picking up the telephone while using the internet during the 90s? :)

Re:

[Meski]

Well, if it was music from the Von Neumann Suite, and there were 2 of them, I'd expect them to mate.

Re:Audio jack to get a standard connector?

[SuricouRaven]

Because apps don't have access to the USB interface. At the very least you'd have to root/jailbreak the phone.

Re:

[agent_vee]

Android 3.1 and up allow apps to use USB Host mode to access USB devices. So you can use game pads and such without rooting. For iOS devices, redpark makes a RS232 cable that plugs into the dock connector. I have used it in a few projects and it works well for sending small data commands but not as a data link for something like tethering.

Re:

[sgt scrub]

In addition to ewanm89's informative answer, it is a building block towards voice commands.

Re:

[roc97007]

> Why not use micro-USB instead of the audio connector?

Because it's an ipod.

Not new

[ewanm89]

You mean like ham operators have been doing to control their SDR radio units for years?

Re:

[Anonymous Coward]

It is innovative if you do it with an ipad2 or iphone4S.

Re:Not new

[Lisias]

1) Take any old but effective idea from the past
2) Use it on a Apple device
3) ...
4) PROFIT!

Serious, nice idea but far, far away from an advance or breakthrough.

Re:

[Anonymous Coward]

You forgot:
1.5) Patent it
2.5) Sue anyone else using it

How to Troll with Patents

[JoeMerchant]

1) Take any old but effective idea from the past
2) Use it on a Apple device
3) ...
4) PROFIT!

You forgot:
2.5) Sue anyone else using it

That would be 3.5, 3) is: wait for someone else to make a significant profit in an infringing area

Re:Not new

[JoeMerchant]

The implied innovation (that keeps coming up in these "iPhone and other smartphone") is in dusting off old ideas that can be implemented with the "now ubiquitous" technology.

Old ideas that require a kilometers long particle accelerator, or three mules and 45 feet of leather strapping aren't nearly as exciting as seeing something cool done with this cheap computer with a radio transmitter that you keep in a pocket next to your gonads.

I once got kudos from Mr. Dean Kamen [dekaresearch.com] for employing a Handspring Visor [wikipedia.org] to do medical datalogging. Yeah, medical datalogging had been done before, our company had been doing mostly that for the previous 25 years, and we didn't have anything to do with development of the Palm Pilot or the Visor, but we recognized the newly affordable, highly portable computers for their potential as a significant component of what was previously a much larger, more expensive, and less portable system. Sure, it could have been done 3 years earlier with "off the shelf" tech, but using the Visor dropped the development costs of the overall system by nearly an order of magnitude. I was a little embarrassed when he said it, but, looking back, we had the idea fairly far along in development as an accessory to interface to the serial port of a Palm Pilot before the Visor was announced, and when the Visor was announced, we backed up (maybe a month's work) to redo the device as an "on the bus" expansion board for the Visor instead. Our timing to take advantage of the Visor launch couldn't have been better - completely accidental, but that's how it worked out.

For what it's worth, investment bankers took over the spinoff company that developed the idea [pdacortex.com], they got all queasy about depending on other companies and "non commodity technology" to support their investment so they went much more vertically integrated, building their own PDA, and recoding all the PC side software in Visual Studio and MFC (from Borland's OWL [wikipedia.org]). In some senses, they were right, Borland and Palm/Handspring did die fairly soon thereafter, but in another view, their prescience about these problems is what hobbled their growth, taking almost a year to re-code the software, and longer to build their own PDA - if they had pushed harder on what they had in-hand, their time to market would have been dramatically reduced, and maybe they would have done better for the original investors. As it turned out, they just plodded along, slipping into chapter 11 [businessweek.com] about 10 years down the road.

Re:

[oakgrove]

But this isn't the first time it's been done with a smartphone. People were making little robots with Android G1 phones a while back using the headphone jack adapter. Is this somehow novel because it was done with an iPhone?

Re:

[JoeMerchant]

But this isn't the first time it's been done with a smartphone. People were making little robots with Android G1 phones a while back using the headphone jack adapter. Is this somehow novel because it was done with an iPhone?

Of course not, beyond the fact that the Android G1 wasn't as media friendly as the iPhone... Android G1, sounds like an obscure geek toy (even though it's not, really). iPhone, well, that's accessible, isn't it? Even my acquaintances in marketing [dilbert.com] have iPhones, and know how to install apps on them too.

I feel the same way when people get excited about a project because "it's so accessible, it's on an Arduino."

Re:

[Anonymous Coward]

Hi, I am the person who did this with a g1. The source and schematics have been available at http://hackaday.com/2010/11/10/android-talks-pulsewave/ for a little longer than a year now. We've been selling these things since March 2010. No slashdot for us?

Re:

[JoeMerchant]

Hi, I am the person who did this with a g1. The source and schematics have been available at http://hackaday.com/2010/11/10/android-talks-pulsewave/ [hackaday.com] for a little longer than a year now. We've been selling these things since March 2010. No slashdot for us?

Submit your story, the gods of what gets posted seem to have gone really soft lately, lots of first time submitters getting published lately.

Re:

[niw3]

So when you read "smart phone or iPod touch", what you understand is "Apple device". Wow.

Re:

[icebraining]

Or like any acoustically coupled modem.

Re:

[ewanm89]

well, not quite, as I'm seriously doubt they are putting speaker to microphone here, no more like a cable direct from headphone jack to microphone jack. Same concept but no accoustic coupling technically speaking and a lot less interference/things that could go wrong.

Re:

[adolf]

well, not quite, as I'm seriously doubt they are putting speaker to microphone here, no more like a cable direct from headphone jack to microphone jack. Same concept but no accoustic coupling technically speaking and a lot less interference/things that could go wrong.

well, technically speaking, what would it be called?

I'm seriously doubt they'd just call it a "modem [wikipedia.org]," no more like obvious.

(also. How is babby formed? [dagobah.net])

Re:

[YouDieAtTheEnd]

It's just a wired connection to a microcontroller. It's not like they're going to be using the soundcard to ouput anything complex, just the necessary control signals to tell the controller to move the motors.

Re:

[adolf]

Which differs from the definition of a modem...how, exactly?

Re:

[YouDieAtTheEnd]

In that it doesn't DEModulate the signals before communicating them to the IC nor would it be MOdulating any signals to send to the phone's soundcard as all of the sensors would be on the phone itself unless you were doing something with tactile feedback. Essentially we're just talking about a wire connected to the controller.

Re:

[ewanm89]

well, technically it's one way, the soundcard on the phone is modulating and therefore is just a modulator, and the one in the robot is demodulating and therefore just a demodulator. Of course, if one hooked up a data channel the other way too, then yes, they are modems. whether they are trying to talk the V.34 or V.92 protocols is another matter.

Re:

[BitZtream]

This actually just a directly connected modem, no pesky air involved in the transmission.

Re:

[oakgrove]

I remember seeing this done with the T-Mobile G1 android phone when it first came out and probably with other phones before then. Is this "innovation" because it is an iPhone?

Everything old is new again

[Anonymous Coward]

The revenge of the modem. How long until someone shows off mad whistling skillz to control this robot?

Re:

[Locutus]

isn't it pathetic that after all these years of having handheld computers(Palm,WinCe, and Linux Zaurus) with I/O connectors being used for all sorts of things our "smart" phones have nothing standard but a audio jack.

Besides reminding me of some of the Zaurus based robots around 10 years ago I'm also wondering why not use bluetooth between the phone and robot and wifi for CC(command and control)?

here's one of the old Zaurus based bots:
http://robotbox.net/project/dahlag/zaurbot

LoB

News at 10

[Haedrian]

Someone discovers abstraction

Not me I guess

[hipp5]

Also, the controls are through the headphone jack. That simplifies the design and means that the robot doesn't need to be linked with only one brand of smart phone.

My smart phone doesn't have an audio jack. Everything is done through the USB port.

Fetch me my slippers!

[YouDieAtTheEnd]

Not sure why everyone is getting so excited about the controller interface being through the audio port since that is old hat and hardly the most interesting connotation of this idea. I'd be much more excited about the possibilities of having a ready built platform with camera, gps, wifi, bluetooth, speach recognition attached to physical actuators. If you were smart enough about the design of the app or provided enough of an interface so that you could program your own behaviors, this could be truly revo

Re:

[YouDieAtTheEnd]

*hopefully it would have spellchecking too

Re:

[JBMcB]

It was a popular way to control toy robots in the 80's. I had an Omnibot that operated entirely on something resembling DTMF tones. The remote control just piped the tones over an AM channel. You could "program" the robot using it's on-board tape recorder, which just recorded the tones it received from the remote. One of the smaller 'brothers' of the Omnibot could be controlled by whistling different notes, or clapping. Same idea.

Re:

[skids]

If you can get a smartphone's USB jack to work in host mode, a hacked deskjet would provide plenty of actuators in a single package.

Dumb

[Kludge]

This is also known as "I'm too dumb to buy a smart phone with a standard USB connector."

Re:

[Relyx]

Unfortunately it's not just cheap, trashy, poorly thought-out phones that have non-standard connectors. Even the iPhone and iPad have a non-standard port. Using the headphone socket is a very pragmatic solution which allows one to cover a very broad range of devices.

Re:

[Hognoxious]

Do more modern phones have the ability to act as a USB host? Most portable devices I've owned/seen just turn their own brains off and mount as mass storage, with the other machine (usually a "proper" computer) acting as the host.

Re:

[icebraining]

The Galaxy S II has OTG support, meaning it can act both as a host and as a slave, although apparently only a few devices (like mass storage drives with a FAT fs) seem to work: https://www.youtube.com/watch?v=giJXF5pIITc [youtube.com]

Re:

[JBMcB]

Archos devices can act as USB hosts for some devices, a holdover from their ability to act as PVRs. Unfortunately the Archos firmware release cycle is awful, and it's broken in every other release.

Re:

[Hognoxious]

Yeah, I have an Archos PMA430[1], which is the difference between "most" and "all". Haven't really used that function much - less than I expected, to be honest.

I remember trying to use the photo app to do a slideshow but it wouldn't navigate to an external drive. It's also fussy about its diet. It recognises thumb drives and if I plug in my multiformat card reader it works with an SD card but not a mini or micro one in an adapter. It has a shell program and it was a lot more fun playing with the command

Re:

[Anonymous Coward]

http://store.diydrones.com/PhoneDrone_Board_p/br-phonedrone.htm
http://www.sparkfun.com/products/10748

http://developer.android.com/guide/topics/usb/adk.html

Almost all android phones can do this in one form or another. My robotics team (sd-ram.us) has been working on building robotic submarines using this technique.

Re:

[Kludge]

Even the iPhone and iPad have a non-standard port.

I know. Therefore they fall in the "dumb" category.

Re:

[roc97007]

Bingo.

Re:

[Migraineman]

DTMF would be an obvious choice, as both silicon and software are readily available. Oh, and the intermodulation issues were resolved decades ago. Doing something custom might be an interesting academic exercise, but it won't gain you an advantage in this application.

soft modem and arduino?

[dns_server]

This is not new so i wonder why it is news, soft modems have been around for a while.
arduino + soft modem + stepper motors.

I was strangely about to start playing around with this when i refreshed slashdot.
I have gotten a soft modem from http://www.sparkfun.com/products/10331 [sparkfun.com] so can i have $92,684?

Re:

[Yvan256]

Troll parsing failed.

Please insert coin to continue.

This could be useful

[davidbrit2]

Siri, open the pod bay doors.

Before you get too excited...

[bradgoodman]

Apple's developer agreement prohibits applications which execute any kind of interpreted, or downloaded code.

So if you're envisioning an NTX-G style of robotics environment for the iPhone, you may need to think again.

Even a simple "Big-Trak"-type (See: http://en.wikipedia.org/wiki/Big_Trak [wikipedia.org] ) application would technically violate such terms. So - Android would clearly be the superior platform of this type of development.

Re:

[matthiasvegh]

So - Android would clearly be the superior platform for development.

FTFY

Re:

[Anonymous Coward]

No, you are not correct.

The Dev Agreement forbids downloadable code, but it does not prevent interpreted code -- there was a small time when interpreted code was not allowed to call iOS APIs directly, but that restiction has been gone for a while now.

Lua is used widely in games on iOS, both directly and though SDKS like Corona [google.com]. There is Wax [github.com] a framework for writing Cocoa touch apps in Lua. In the App Store itself you will find 4-5 different Basic interpreters, a python interpreter, ruby, and the excellent

Re:

[account_deleted]

Comment removed based on user account deletion

Re:

[jockm]

For the love of... have you bothered to actually read the agreement? Citing from wikipedia:

3.3.2 — An Application may not itself install or launch other executable code by any means, including without limitation through the use of a plug-in architecture, calling other frameworks, other APIs or otherwise. No interpreted code may be downloaded or used in an Application except for code that is interpreted and run by Apple’s Documented APIs and built-in interpreter(s).

So yes you can embed an interp

Re:

[account_deleted]

Comment removed based on user account deletion

Re:

[BitZtream]

While your point is valid as the rules are written, a power point viewer or RDP client can be technically considered a violation of those rules as they receive commands from a file or network connection telling them what to do.

That and there are other devices that do this already.

You can buy an remote car that works these way from the local bestbuy, download the app to control it from the app store. It doesn't use a direct wired connection, it has a transmitter that plugs into the headphone port, but none

Re:

[jockm]

No skulduggery needed, see my other post to bradgoodman (I wasn't logged in, so it is an AC), but programming environments are allowed on iOS. There are fair number of examples in the App Store already

Re:

[psydeshow]

Even a simple "Big-Trak"-type (See: http://en.wikipedia.org/wiki/Big_Trak [wikipedia.org] ) application would technically violate such terms.

OMG the Big Trak! Somebody needs to bring that thing back so I can finally own one.

Why wired?

[Iamthecheese]

It would be better to control the robot via 802.11

Re:

[WrecklessSandwich]

It would be better to control the robot via 802.11

That requires an 802.11 network as well as 802.11 hardware on your robot, which costs significantly more than the onboard microcontroller.

Re:

[Locutus]

bluetooth then since standard USB is not an option here in 2011.

LoB

Re:

[roc97007]

I would guess, wired makes it more hack proof?

Smartphone = Universal Remote

[Sentrion]

Expect to see more stories like these in the future. Before long, all of your old appliances that had displays and controls will now be nothing more than a box with a logo. All communication will be via blue tooth and wifi, and your smartphone or other compatible device will be the interface and display. The controller inside your appliance will probably be the same type of microprocessor used in your smartphone. Even classic "dumb" devices will soon have this capability, so you will be able to walk up to just about anything, hold your smartphone up to it, and see what it is, what it is doing, how it is doing it, and what changes you can make to it.

Prior art

[fikx]

http://www.cs.cmu.edu/~pprk/ [cmu.edu]
back in the days of Palm computing, it was known that letting your portable devices run around from time to time on their own was good for their health... :)

No need for an R2-D2 accessibility law now.

[__aasehi2499]

I never thought I would see Star Wars tech come to reality, that thing always looked like a phone jack to me.

You know...

[roc97007]

...that this is going to be used in a summer popcorn flick, probably starring Bruce Willis or Will Smith.