Hacker Newsnew | past | comments | ask | show | jobs | submitlogin
Meet The Double (YC S12), A Teleconferencing Robot With An iPad For A Face (techcrunch.com)
107 points by davidcann on Aug 13, 2012 | hide | past | favorite | 66 comments


That's a pretty impressive price point and I'd love to know more about the in house manufacturing.

The one company that has really succeeded in this area is Giraff. They've done so by intensely focusing on in-home care and the user experience for the person interacting with the robot. Things like end user controls to approve/reject someone who wants to start driving the robot.

Anybots, on the other hand, focused more on the driving experience. Their rendered, in-browser, drive window helps overcome the awful user experience of trying to turn under lag. They also did a lot of hard work to solve wifi roaming and seamless handoff.

I wish Double the best of luck with my advice being that you need months of uptime and recorded daily use in a target vertical before you're ready to release.


Thanks! We're out in Mountain View now, but we built out our own shop a year ago in Miami, FL. About 90% of the manufacturing on Double is still done there today by our contract workers. We have a Haas VF-2SS mill and have made custom fixtures for most of the parts on the robot. We purchased a Stratasys FDM printer and brought it with us out to CA with us to do rapid prototyping during YC. All of the prototype electronics were hand-soldered under a microscope and tested in house, but we will soon be handing that off to a PCB Assembly house now that the boards are finalized.

As far as driving goes, we have 3 or 4 different methods we want to experiment with once things settle down, but so far the joystick-style is most intuitive. OpenTok's new iOS to iOS video has cut down on the lag a bunch.

Thanks for the comments!

-marc @ double


Very cool, I hope you have a Dimension series, those are sweet.


I'm not really impressed with the price. It's just an iPad on wheels, and you have to provide the iPad.


Try buying your own lithium ion battery that'll last eight hours then. Companies need to make a profit too...


We used a pair of VGo telepresence robots at our NYC and Philadelphia offices last year. Here's some video from the demo, just before we got them: http://www.youtube.com/watch?v=L3tLLzOqKs4

I found the VGo to be a huge help when it came to being present in an office that I couldn't physically go to at the time. I was the head of a department that had groups in different cities, and it was a lot easier for me to feel I had a cohesive team when I was able to actually "be" there instead of just on Skype. It's hard to overstate the value of being able to go join a quick work session at the couches, or go looking for someone in an office where people are always busy and wandering around.

Of course, in a large company where not everyone is an early adopter, to put it politely, there's also an annoying side to having a telepresence robot. Some people spend more time focusing on the novelty than getting to work, and it can be a distraction. A lot of people weren't comfortable using the VGo themselves, even though it was available for anyone. There are definite social hurdles to overcome with some people.

I'd probably like the Double even more than the VGo, because it looks less intrusive, has a larger screen thanks to the iPad, and the adjustable height is a very big deal. I'm looking forward to having one in the future. Love the price, too.


Interesting perspective as the remote user, but I wonder how your employees felt about that setup. It was easier for you to feel you had a cohesive team, but did they feel more cohesive for having a robotic you floating around?


So their biggest competitor is Anybots, but they're funded (through YC) by Anybot's founder and going through YC in the same physical space as Anybots?

It strikes me that the main axis of competition is simply price. Otherwise it just seems like co-opetition.


I'm no longer running Anybots, which has moved to new premises. I like both products for different applications.


Do you have another startup in the works, or are you concentrating on YC full-time?


Direct link to our site: http://www.doublerobotics.com


The site says it charges with a wall adapter. This may be a silly question, but can someone driving the Double manoeuvre it into "charging position", or do you need fingers to plug it in?


For the reason that this is real technology and not just another consumer/social/we'll figure out the business model later idea, I hope this grows and becomes huge, or at the very least spawns some other applications.

In a week that NASA lands a robot on Mars, I'm glad to see real technology being created, that perhaps looks crazy or inapplicable at first. Great stuff guys.


The hardest part for telepresence robots is finding a market/use case to grow in. Being cheaper is a great first step, but who are you selling to?


I work remotely for CloudFlare. I wish we had one of these so that I could attend meetings at the company or hang out at someone's desk.


Right, but how will this be better than screensharing and teleconferencing? I'm not trashing the technology - it is an extremely well done version of a telepresence robot, combined with a spectacular price point. I'm worried that they're not competing with who they think they are competing with.


> how will this be better than screensharing and teleconferencing?

We have a remote worker. During meetings, the iPad that he's on just sits on a desk, pointing at only some of us. It would be nice if he could turn around and face the person he's addressing. It's a small thing, but important.


The troll solution would be to buy an old turntable and put his iPad on that. No more manual turning!

Seriously, an app that can remotely command a turntable/stand would be neat.

TouchOSC would be able to do it, provided the right proxy and hardware at the other end. http://itunes.apple.com/us/app/touchosc/id288120394?mt=8


Not a bad idea, I have a USB missile launcher but I've been having a hell of a time talking to it via USB. :/


It means I can go into a meeting and 'sit' at a table with people and turn and see what's being written on the whiteboard.

It's much better for me (because I can attend) and it's better than screensharing and teleconferencing because the participants in SF do not have to use some program to accommodate me.


> and it's better than screensharing and teleconferencing because the participants in SF do not have to use some program to accommodate me.

This is huge. There's nothing worse than holding up an entire meeting while someone struggles to connect to you via video chat while IT is in there sweating bullets because a new driver/update is trying to install itself.

Removing this burden from others would be worth it to me.


This isn't an argument for a telepresence robot - it is an argument for a Cisco teleoffice installation. Having a robot adds a whole new layer of complexity and issues.


Then after the official meeting, when attendees go out to the food table in another room and the tele-attending people can't digitally follow them, I'm sure the attendees will all just not discuss anything important, right?

Or "hey, Bob, I'm glad you could use the teleoffice installation to join us today. Gosh, I wish I could have you with me tomorrow when I go across town to check out a small factory. You know so much more about that stuff than me, but unfortunately the tele-office is huge and all."


First, I think you're overestimating how easily people will adapt to having these robots around. I hope there's no stairs or elevators in between the official meeting and the food table.

Second, does this small factory have high speed wifi? In my experience, that's very rare.

You know what does work in these situations? Flying out to the meeting.

As a technologist and possible early adopter, it is really easy to see how YOU'D use something. It is far harder to put yourself in the shoes of another and judge it from their perspective.

That said, once telepresence robots are in place, they seem to be very sticky - most people don't want to give them up. [1]

A key quote from [1]: "the greatest barrier keeping telerobotics from conquering the office probably isn’t hardware or software, it’s red tape"

[1]: http://singularityhub.com/2010/08/23/exclusive-head-of-robod...


I'm not sure why you're being down-voted, your response was on topic and adds to the discussion. Methinks some folks might need to brush up on their HN etiquette.

Anyway, my argument on its own probably isn't sufficient to justify running out and buying a $2000 robot + iPad, etc. but it is surely _part_ of bigger argument for these things.

Also, in response to your later comment about how well-recieved these units will be, I think they will be ubiquitous in a few years. I think you'll see them used in all of the areas discussed so far plus many more we haven't even though of. I could see an army of these things in stores, checking with customers, providing information at airports. The uses are nigh-endless.

Remember, just a few years ago the idea of having a _phone_ that went with you was crazy talk. Now it would be weird to _not_ have one.


I would love for these to take off - a telepresence robot in every meeting would be grand.

Ubiquity requires more than just the robot though - it requires real policy change and upsetting the status quo. I think that's the hardest part.

Also, RE: downvotes. It is far easier to downvote than it is to engage in a rational discussion. So, thank you for that tankbot. =)


Except they have to use a _robot_ to accommodate you - hardly less invasive than some software. =/


How about during surgeries where a remote doctor who is highly trained can wheel around (virtually) and instruct or guide a less experienced surgeon?


According to the article, they want to do it "the lean startup way".

So... why does it have to be self-balancing? Just add a third wheel and it's always stable - no need for gyros and complicated electronics. And as a by-product, you have immediately lowered the price massively: it's just a few motors then instead of that balancing magic. My price lowering guess would be between 50% and 80% lower. That would be really "the lean startup way" in my book.

(Idea courtesy of http://maddox.xmission.com/c.cgi?u=segway_more_complicated_t...)


Thanks for your thoughts, but the third wheel theory isn't actually true. We investigated this, of course, and a third wheel without balancing provides a pretty crummy and bumpy user experience. The key is reaching the magic human height while keeping the base no larger than a human's foot prints. Look at the VGo, they opted to limit their height to about 3' and add a third wheel, but the driver is always looking up at people's stomachs and up people's noses. It's awkward in a standing conversation and, even at that height, it's a bit shaky riding around - not to mention the extra counterweight they need to put in the base to weigh it down. Two wheeled, self balancing is simply a more elegant solution all around.

Also the extra cost of the accelerometer and gyro is almost negligible these days, thanks to smartphones requiring them to be made in such high volume. For both together, it's maybe $20 or less in volume.


Interesting, that clears up my reservations, thanks!

I feel like the biggest issues might be missing ambient spatial awareness for the user - with that non-wideangle lens and mono microphone you can't see the whole environment at once, you have no idea where somebody is talking from, and you cannot turn quickly enough to always track what's happening in a room, if people are leaving/entering etc.. How do you plan to make the user feel more "there"?

Also, micro-managing the movement of the bot would take up a lot of time that I don't have to spend when I'm present locally. E.g. when I'm present myself, I just look somewhere, then think "I want to go there" and it all happens automatically. With the bot, I have to manually navigate around plants and desks. It would be great if we could do the same thing with the bot: just touch on the screen where you want to go, and the bot figures out the rest. Or show a floor plan, touch where I want to go and the bot does it. Do you have plans in this direction?


Ah, I'm glad you see the potential of where this technology can go! Those are all great ideas. "We have nothing to announce at this time."


Yep. Bring it down to $500 and I'll buy one!


I'd find this difficult to tolerate at work. I can't imagine anybody addressing me with one of these. Not only that I am certain no work place I have worked at would ever buy something like this. And they're all happy to spend thousands of dollars on things which would improve our productivity.

I was ready to dismiss the idea until I saw the end of the video. I think outside of business/work it could be quite useful (as demonstrated by the art gallery).


The art gallery idea struck me as interesting at first.

On second thought, however, why would you want to look at art through a low quality iPad camera when you could browse a website with high-res images of the same stuff?

I'm not sure that usage would pass the gimmick test after the novelty wore off.


Another complication is navigating through the museum without violating the personal space of the visitors. I guess they could only be allowed in the after hours but then how do you prevent the experience from deteriorating into high-tech bunch bumper cars.


That's really neat, I think this has a number of uses and not just at work.

How does it stand up though? Is there an acceleramoter or similar powering a balancing act?


Yeah, probably a few gyroscopes/accelerometers.

I would guess that it is a beefed up version of something like this: http://blog.makezine.com/2012/03/28/an-inexpensive-self-bala...

I like their tiny kickstand as a solution for keeping the thing upright without draining the battery. I wonder if that deploys only when the bot is in sleep mode, or if it can come out when someone is using it but "standing" in a single spot for a prolonged period of time.


Yes, that's right! Our very first prototype was using a 5 degree of freedom IMU board from SparkFun.

Yes, the kickstands can be deployed manually during a call or automatically, if the robot is sitting in one place for a long time. When the user attempts to drive again, the kickstands instantly lift off the ground and the robot can drive again.


I had a related idea recently: Just hire a tall actress to carry around an iPad running Facetime. (Perhaps with a custom shoulder-strap?) It's going to be a bit longer before robots get up to the level of humans in terms of versatility, social awareness, and attractiveness. Until then, using a human would have certain advantages for a high-end solution.


I'm excited to see some hardware concepts coming out of YC in the past couple of years. I think that the lean startup model is starting to be much more viable (vs 5 years ago) in the hardware space thanks to low cost prototyping electronics, 3D printing, and small batch size manufacturing.


There are a bunch of hardware startups in this batch. I'm not sure why. Could be just random variation. But for whatever reason it's kind of the hardware batch.


That's great! Looking forward to hearing more about the others after demo day rolls around.


Impressive design, both mechanical and graphic. Does anyone know who produced the video?


Thanks! It was a collaboration between missioncm.com and cc-prod.com.


I built something like this 2 years ago at iOSDevCamp out of legos. http://zbowling.github.com/NXTRover/

I love that of this was taped at TokBox :-)


Nice, I love it. We started out super simple and just kept improving it and making it taller.


Check out the prototype, made public at WWDC over a year ago: http://www.youtube.com/watch?v=2BXM9PYaXY0


Haha, yes, we've come a long way since then!


For those who are interested, Suitable Technologies (from Willow Garage) will be making an announcement next month:

http://suitabletech.com


Here's another competitor... looks like the home telepresense market might be ripe and ready! http://9thsense.com/


Another competitor designed for industrial settings: http://collaborate.io

Disclaimer: I'm the founder


That looks pretty cool. It looks like you nailed the precise zoom feature, which is key for your use-case.


>> It uses OpenTalk to essentially let you video chat with those people.

What is "OpenTalk"? I've never heard of that protocol/API/software before. And Googling turns up nothing except some abandonware from 2009. http://download.cnet.com/OpenTalk/3000-2150_4-10568159.html


It's actually OpenTok. Check it out here: http://tokbox.com/opentok/api/documentation


Thanks, I asked them to fix the typo in the article.


Opentalk was also once the name of a Smalltalk distributed object framework similar to DRb. (Distributed Ruby.)


Does the iPad need to be charged seperately, or can it recharge when the robot is plugged in?


Yes, to keep things simple, the iPad and robot communicate wirelessly, so the iPad is currently charged separately.


They should be integrated. You need to be able to 'park' this thing on a charging pad and know that it will be there, fully charged when you need it. Otherwise you're creating more work for someone at your remote site (assuming there is someone there to create work for, these could probably be used at unmanned sites with great success).


Agreed and noted. "We have nothing to announce at this time" is our standard response in a situation like this.


Cool concept, hope it does well.

Seems ironic that "They were actually developing a product that required partnerships with Chinese manufacturers, and they found themselves increasingly frustrated by the fact that they couldn’t keep an eye on the project".


This is pretty darn cool and pretty darn freaky at the same time. It may want to use both front and back cameras to see if someone is coming at it from the back since it won't be able to get back up on its own. :o)


So, it's a Segway with FaceTime? The main question is why? Even the video can't seem to find a decent use case. The art gallery idea is ridiculous, why would you physically control a robot that wanders an art gallery when the gallery has a website? Also have you ever tried actually wandering around a gallery? The robot wouldn't stand a chance. What happens when someone knocks it over?

In the workplace this would just be embarrassing. Not to mention a security threat, there's no way any serious corporation would allow roaming cameras in their offices.

If I put my idealist hat on and try to imagine the future with these, I keep coming back to those Segway videos of the hype of a new kind of city, a new kind of transportation.

This really is a technology in search of a use case, that isn't already being addressed. I read comments that say they would use these to attend meetings, how does this robot compare to Cisco's video conferencing, or FaceTime or even just calling in from the car/home. At work we use screen sharing and audio/visual conferencing which seems to work just fine, and where it doesn't it will evolve.

I'm wondering how this robot would push the elevator button to get to a meeting on another floor? There are just so many problems with this idea, not the least of which is who would actually buy one.


It's been useful already. One of the founders was out of town at the last YC dinner and but he participated in a conversation in the lobby with me and his cofounder through a Double. He wouldn't have been able to if he'd been using a standard teleconferencing setup, because it wasn't a pre-arranged meeting.

In my experience, many if not most of the most important conversations are not pre-arranged meetings. If you use existing teleconferencing technology you miss them all, and if you use a Double you don't. That's a qualitative change.


Pre-arranged meetings are the first thing that jumps into people's minds, but do telecommuters really want to replicate the experience of being in meetings? For me, one of the nicer aspects of telecommuting is I can call into meetings and during portions of the meeting that don't apply to me (or are boring) I can work (or surf the web).

On the other hand, people do have a love/hate relationship with meetings: http://online.wsj.com/article/SB120519462810425677.html

I can buy into the importance of impromptu meetings. I wonder if the Double team can figure out a way to market around impromptu vs pre-arranged meetings and love/hate meetings without alienating too many people.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: