Hacker News new | comments | show | ask | jobs | submit login
BlueRobotics – T100 Underwater Thruster (bluerobotics.com)
140 points by slackpad on Aug 12, 2014 | hide | past | web | favorite | 58 comments



I've experimented with building a submersible robot and - by far - the most challenging aspect of the build was creating an interface between the motor and the propellor that was waterproof in a reliable way. (applying grease or something hydrophobic on a shaft lasts only so long)

By abstracting away that schlep for me, - a schlep, which is usually the least interesting bit about a machine - these guys might be able to create something analogous to Arduino for marine robots. Before Arduino, you needed a complete test bed to get a micro-controller up and running. After Arduino, you usually focus on what you want to do, instead of getting the micro-controller to work. That shift makes creating something extremely accessible, allowing anyone to jump in and try things out. A line of products like these, might actually make robotics far more accessible than any number of kits before it. In other words, this could be the start of something incredible.

Edit: I just realised another way of phrasing what they're doing. Imagine how hard payments was before Stripe came along. These people are basically the Stripe for marine robotics, which is really awesome.


I saw a really novel water tight propeller system where the shaft of the propeller had permanent magnets on it, and inside the boat there were coils that pushed against them. The propeller was in a hermetically sealed shaft. No crossing points at all. Probably not as efficient as other solutions but I am sure it did not leak anywhere.


That's actually quite impressive and it ties into why I started the project. When I was a teenager I read a paper that talked about the drive used by Captain Nemo in Twenty Thousand Leagues Under the Sea. The magneto-hydrodynamic drive† basically tries to push seawater through a set of coils and, as you've said, it's very, very inefficient and currently impractical.

Reading that paper inspired me to look into this and I ended up trying to make a submersible with the limited budget I had. It was very frustrating and I just gave up and ended up trying to make linear actuators for touch sensing instead.

http://en.wikipedia.org/wiki/Magnetohydrodynamic_drive

http://www.skewsme.com/mhd.html#axzz3ACp7NENa


Wasn't this also the caterpillar drive from "Hunt for Red October"?


Only in the movie. In the book it was an impeller.


Yup!


Wow, I never a Magnetohydrodynamic_drive existed! Has there been any attempt to use it in air instead of water?

I guess you'd just need a really high voltage?


> the most challenging aspect of the build was creating an interface between the motor and the propellor that was waterproof in a reliable way.

There does not have to be interface. Put stator inside and rotor outside, aquarium pumps works on similar principle.


The losses across that gap are expensive, though. Aquarium impellers are moving water at very low pressures.

It works if the isolation layer is very thin. However, that prevents you from going deep because of the pressure differential. Now, I wonder if you filled the sub with something inert and kept the layer thin, then the pressure differential would be a lot less and the layer a lot less likely to rip open. Might work.


> By abstracting away that schlep for me...

> I just realised another way of phrasing what they're doing. Imagine how hard payments was before Stripe came along. These people are basically the Stripe for marine robotics

Hey, didn't you get the "schlep" and "Stripe" part from [0]? ;).

[0] - http://paulgraham.com/schlep.html


Excuse my ignorance, but can't you use the kind of seals that are used at the end of engine camshafts and crankshafts?


A similar company is SeaBotix [0] they have a similar [1] thruster for sale but it uses a brushed motor. They have a newer generation product (not separately listed) that uses brushless motors that has integrated motor controller so all you need to do is send it power and I2C speed commands.

For people interested in this I recommend checking out the RobotSub [2] competition. Each team releases papers allowing you to see how the whole vehicle fits together.

0 - http://www.seabotix.com/

1 - http://www.seabotix.com/products/auv_thrusters.htm

2 - http://www.auvsifoundation.org/foundation/competitions/robos...


I did RoboSub for 4 years (probably the best thing a student can do to become a better engineer. That's another story for another day). The only issue with the SeaBotix thrusters is that you have to take off a screw and inject grease into them every year or so.

No indication of how BlueRobotics' thruster actually fixes this problem other than the fact that they claim they fixed it.


Something I wanted to ask for a long time:

> Many of the T100′s components are American made and sourced. It’s assembled in the USA. We’re proud of that.

Is that "made in the USA" thing really such a big selling point in the US? Because to me as an European that really is a little alienating. A "made in the US" logo - ok. But pointing out the proudness in marketing materials? Strange.


There's a feeling that manufacturing jobs are leaving the US at a quick rate, that anything we can do to keep them here is a 'good thing', and supporting companies that keep jobs here is equally good. Marketing as such is therefore just trying to take advantage of that.


Where are you from? I'm in Germany right now and the "Made in Germany" thing is getting on my nerves - you get cheap crap made in sweatshops by Turks, but hey, it's Made IN Germany, whoohoo...


It is, oddly enough.

It's interesting that this particular product would make use of the gimmick, however, given it's potential international market.


Another interesting alternative to a traditional motor is the "Slocum Glider", which dives and propels itself through the water by pumping oil between its body and an external bladder.

http://www.whoi.edu/main/slocum-glider

http://en.wikipedia.org/wiki/Underwater_glider


I've used thrusters in the $1000+ range and they tend to need maintenance every several hundred hours of use in nearly ideal situations. This isn't an easy project and can be very difficult for student groups (for example) to get right or to get affordable. Good luck to this project.


Wow this looks amazing.

I feel left out in the coming 'robotics revolution'. I am planning to take a couple of years off from work and switch careers into something robotics. I've been building my math base past couple of months, but I am utterly clueless as to where to even start with robotics.


One of the things that I've always enjoyed about robotics is that it is a 'systems' discipline. It employs mechanical, electrical, and software components into a complete system. My experience after 25 years of robotics club meetings is that there are few people who can 'do it all' and that means you pretty much have to team up with people who have complementary skill sets.

So depending on where you find yourself most comfortable consider focusing on that part of the problem. Could be algorithms (software focus), could be mechanics, could be 'mechatronics' (which is the term the local community college uses for a combined electronics/mechanical engineering sort of curriculum).


For software - OpenCV, ROS (robot operating system) are good places to start. Unfortunately most robot research in universities suffer from a very bad 'not invented here' syndrome, so a lot of wheels are reinvented over and over. It has been about 5 years since I seriously payed attention to the software ecosystem so this might be out of date.

For embedded hardware, I'm a big fan of the BeagleBone Black. People talk a lot about the Raspberry Pi, but the BeagleBone makes a bit more sense. Look at the OpenROV project as a good example of that.

Lower level, there is a lot of stuff related to feedback controlled (closed loop) motor controllers that is key.

Higher level, pathfinding AI is interesting. Even things like inverse kinematics, thought that is largely a solved problem.

Some of my favorite books are Probabilistic Robotics for the theoretical stuff, and Robot Builders Bonanza for the really practical hobbyist stuff.


Another great hardware platform, the Jetson TK1. Much more capable than the Pi (I can't speak on the BeagleBone), and even though it's new it has great software support (OpenGL, OpenCV, CUDA).

http://elinux.org/Jetson_TK1


> where to even start with robotics

With an EE, CS, or ME degree? Minor in biology. Nature has made by far the best robots. Definitely useful to learn from it.


I once read a fascinating article about a grasshopper(?) that uses such a fast and sudden movement to jump that the only way it could keep its two legs in sync was with a biological "gear", with teeth and everything.

Edit: Found it: http://www.smithsonianmag.com/science-nature/this-insect-has...


100% agree with the nature comment. You'd enjoy this ted talk by Robert Full (if you haven't already seen it) https://www.youtube.com/watch?v=Ascql_RoeBU


If you want to start now and have fun immediately I recommend picking up Lego Mindstorms: http://www.lego.com/en-us/mindstorms/?domainredir=mindstorms...


presumably this? https://www.kickstarter.com/projects/847478159/the-t100-a-ga...

I am worried they allow water running through the assembly, sea water is very corrosive and can't be in contact with any interesting materials (e.g. high performance magnets). /*

/* ok they use plastic bearings and stuff, not sure what the magnets are.


From the kickstarter: "The core of the motor is sealed and protected with an epoxy coating and it uses high-performance plastic bearings in place of steel bearings that rust in saltwater."


still somewhat skeptical that an epoxy seal will last next to turbulent water with sea debris. I know other strategies are magnetic couplings. Still, maybe replacing cheap motors frequently will be cheaper than a long lasting one (if one exists??). Make sure your mounting plates are accessible!


By way of reference, there is nothing comparable within 10x the cost of this thruster. If you're worried about longevity then perhaps $1,000/thruster is more reasonable.


Oh wow, I did not see the price, yeah that's crazy good.


i assume it'd be something better than just a standard epoxy. there's plenty of underwater-specific epoxies and coatings used in the oil&gas industry that'd hold up just fine.


Thanks for the link, so they are using a brushless motor. I really like that solution for water integrity. I would guess they seal the magnets in epoxy. That would make it reasonably durable.

The surfboard is interesting. Having listened to some of the issues the wave-glider ran into in order to operate autonomously on the ocean I expect the same sort of folks will come talk to them. I heard a story (and it could be just that), that one of the wave glider experiments was going to be to count whales along the California coast, the plan was to put an active sonar pinger on the robot and ping once an hour to see if there were any whales nearby. The story goes that the project was shelved when the Navy called and informed them they would not be putting active sonar pingers into US coastal waters.


the Navy called and informed them they would not be putting active sonar pingers into US coastal waters

This makes no sense. Active sonar pingers are on pretty much all boats apart from small dingies.


Here's a project that used a passive listening device on a wave glider:

http://www.fondriest.com/news/wave-glider-robots-listening-t...

The whale monitoring I'm familiar with is all passive listening as well. The navy has a long history of pissing off marine mammal experts by blowing things up underwater, so it makes sense that the Office of Naval Research wouldn't fund active sonar. So your recollection has elements of truth.


Extremely excited about this project. Anyone that wants to get into robotics for the ocean should check out http://www.amazon.com/Underwater-Robotics-Science-Design-Fab...


My senior engineering project in college was the development of a GPS-guided robotic boat for recreational purposes. Our team consisted of three electrical engineers with very little machine design experience (this was just the first of many aspects of this endeavor which were hilariously ill-thought out), so we ended up bashing together a makeshift thruster out of plumbing fittings and an electric trolling motor, with predictably awful results.

Seeing this brought back memories. A self-contained thruster unit intended for integration into a larger robot might just have saved our project, had it been available four years ago.



Looks a lot like what we did for OpenROV. https://www.youtube.com/watch?v=8q4CirrIgWo


Wow, looks great! I don't have much interest in making a boat, but this looks like a great Kickstarter.

Also, the thrust and efficiency looks crazy good on those! At 50% it looks like 1.2kg + what looks like 30g/W, which is insane efficiency compared to multirotor motors on the market now. I'm guessing the fact that you're in water helps substantially with that, right?


It seems like they should have waited for the surfboard thing to make it's journey before launching the kickstarter. It would have made the project seem a lot more credible. Like another poster mentioned, I'm sure a lot of drug traffickers would be willing to invest if it's proven to work


This and Saildrone promise a very exciting future for ocean exploration and data gathering.

http://www.wired.com/2014/02/saildrone/


Can't wait to attach a gopro and make an underwater drone.


I never thought aerial hobby drones were that interesting, but that sounds fascinating, as you'd be looking around in an environment seldom experienced otherwise.


I've wanted to do this and get an Oculus in the mix as well!


After reading this[1] xkcd post about the ROV he built, I've wanted to try and build a similar one when my son is old enough. Then after seeing the Oculus + drone video I wanted to add that in as well and go "diving" in Puget Sound

[1] http://blog.xkcd.com/2010/11/05/submarines/


dude, imagine battle bots...but underwater with Oculus VR.


I am interested, tangentially, in an AR display that would fit into my dive mask. I wonder how hard this would be?


"What's an underwear thruster? ... Oh, wait."


I beat drug smugglers will love to use something like this.


They already have submarines: https://en.wikipedia.org/wiki/Narco-submarine


This would be unmanned and cost a fraction


Interesting idea. Compared to a submarine yes, but what this would more likely be up against, from the perspective of cost is a panga boat:

http://www.independent.com/news/2012/jul/26/13-people-arrest...

"The push out to sea is spurred by increased crackdowns on the border, and pangas loaded with multiple fuel barrels have been spotted as far north as San Francisco. The Los Angeles and Long Beach areas have seen a similar increase in incidents."

Lets see, to ship ~$6 million worth of pot paying someone a small fraction to drive a boat up the coast, plus a few thousand for a boat, I'm not seeing a reason why they'd want to make a drone sub to do this.


Can someone name a successful, on-time, major hardware kickstarter project?


Can someone name a successful, on-time project? You're setting the bar quite high, aren't you?


The Oculus Rift kickstarter seems to have been delivered on-time.


O U Y A




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

Search: