About this project
WHAT IS THIS ANYWAY?
We're asking for your help to produce and test the ROBOTBOAT MARK VI. Each robotboat is fully autonomous, needs no fuel, and will bring to bear myriad sensors at remote points on the surface of oceans, lakes, and rivers. Think of it as a satellite for the seas.
We need lots of low-cost self-propelled robots to study the health of the world's water. Never send a person to do a buoy's job! The real value of this technology lies at the fleet level, but given the extremely high performance required of each boat, purchasing even a few of them is simply beyond our means. Building one boat represents a huge step towards cheap data. Building more than one boat will allow us to test swarming algorithms, implement complex communication schemes, and take greater risks (e.g. undertake longer, more challenging missions).
Bringing this opportunity to the kickstarter community will allow us to develop this technology MUCH FASTER than we could if we have to fund it ourselves; the more resources we can raise here, the faster we can deliver.
WHY SHOULD I CARE?
Perhaps Lilly Tomlin said it best: "I said 'Somebody should do something about that.' Then I realized I am somebody."
No matter where you stand on issues like climate change, marine conservation, offshore energy production etc., we hope you'll agree that everyone is better served by policies and behavior based on empirical scientific evidence. We should all care about how we manage our water resources.
We think Robotboat is the best way yet to generate information about the world's oceans.
- More data = better ocean management
- Fully automatic = cheap to acquire data
- Fast = station-keep in high seas; get to points of interest quickly
- Durable = collect longitudinal data; self-righting
- Each boat = millions upon millions of datapoints
WHO ARE YOU?
Walt, Rob, TJ, and Eamon... Hi.
We're a professional engineering team. Between us we hold 8 patents and have over fifty years of engineering experience.
Our principals (Walt & Rob respectively) founded these companies:
TJ and Eamon have worked for Walt making satellite subsystems for the last few years (among other pursuits). We've designed and executed dozens of successful missions. We've produced space flight hardware, ground station equipment, and remote sensing systems for the most demanding customers in the world, Collectively, our expertise is in developing robust mechatronics.
HOW ARE YOU GOING TO PULL THIS OFF?
We've built & tested 5 autonomous sailing robots over the last several years.
Our boats have completed hundreds of hours of fully autonomous navigation and traversed hundreds of kilometers on the water.
We've proved a robust sailing algorithm. In the GPS trace below you can see the boat complete upwind tacking maneuvers without human control.
We've completed the Robotboat Mark VI design. We're going to build it like a surfboard; it will be incredibly strong and be capable of being rolled in waves indefinitely.
We won't quit.
Neither will the boats. The Robotboat Mark VI features a patent pending self-righting mechanism, the first and only such system in existence. It can prosecute targets day or night, rain or shine, wind or doldrums, thanks to its advanced wing sail and assistive electric propulsion system.
And... we'll need your help. R&D of this nature is not cheap.
- Mechanical and electrical components = $40k
- Tooling (molds & dies) = $30k
- Average cost per test = $1k
- Expected test quantity = 10x
- TOTAL = $80,000
We're trying to build as many of these boats as possible.
As soon as we receive this kickstart, we will execute the following schedule:
As your sponsorship allows us, we will undertake more ambitious missions in the next few months:
Measure oil leaks
Measure the Pacific Garbage Patch
Measure Boundary Currents (e.g. the Gulf Stream)
Measure Marine Life (video & acoustic)
Measure Coral Reefs
Sail Graphics Placement:
The earlier you pledge, the higher your logo ; )
Postcard: (NEW REWARD ALERT)
The details are yet to be finalized, but hopefully the picture above gives you an idea as to what we have in mind.
Black ink on grey shirts emblazoned with ridiculously large prints of a wireframe model of the Robotboat Mk VI (adult unisex sizes available XS - XXL). These will be hand-screened in VA.
Pro Sailing Jacket:
Originally these were just for the four of us, but we decided to make them available to our most generous backers. They will be available in the full range of women's and men's sizes and embroidered with a picture of the Robotboat Mark VI. Though the picture above is just nominal, the jackets will be of extremely high quality. Trust us, we've frozen our tootsies off while testing Robotboats enough to know that good gear is essential when people venture out on the water.
3D Printed Model:
Think bathtub/desktop size...
Special thanks to our international backers! We'll have to ask you to help mitigate the extra shipping costs required to deliver your rewards. For addresses outside the U.S.A. please add $8 for keychain / $15 for tshirt / $20 for 3D model / $35 for jacket to the appropriate rewards levels at right.
Thanks for checking out the Robotboat!!!
We live in a very exciting time in sensor development! Robotboat Mk VI has ample power to run all sorts of electronic sensors. The boats can handle many low-power sensors continuously, and high power sensors on lowered duty cycles. At the most basic level, any vessel that is trying to sail will need to 'know' where it is, where it's going, which direction it's facing, and how the wind is blowing. Additionally, it would need to know where its rudder and sail are, and, practically speaking, lots of other state variables. The baseline sensor package for the MK VI includes GPS, IMUs, cameras, temperature sensors, relative wind direction sensors, and a wind speed sensor. We've built in the 'plumbing' for hydrophones / acoustic modems, even more cameras, hull speed sensors, conductivity sensors, oxygen sensors, barometers, opto-fluidic water quality sensors, hydrocarbon sensors, Geiger counters, salinity sensors, CO2 sensors, pH sensors, pressure sensors, and load cells. We've done our best to future-proof the design by standardizing payload interfaces and can provide power and telemetry to analog, digital, whatever. On-board sensors can be placed at various places along either of the two hulls, or the mast, or the keel. Placement can be high enough for radar and low enough for active sonar profiling and triangulation.
When Walter made the Mk I in 1994, the name sounded a lot less generic. Since then, we've tried to come up with something better a few times, but never could seem to agree. After a while we had tabled the discussion so often that we just gave up. Our time is probably better spent engineering anyway. We may open up the discussion to our backers. BTW we'll likely default to "Robbie" for the Mk VI S/N001 (props if you catch the reference).
First of all, satellites are extremely expensive to produce and deploy (though that is changing rapidly, thanks in no small part to https://directory.eoportal.org/web/eoportal/satellite-missions/c-missions/cubesat-deployer & http://server2.tigerinnovations.com/homepage/news/TigerInnovations_STPSat1_Automated_Operations.pdf). A satellite capable of doing useful science will cost hundreds of millions of dollars over its complete lifecycle. Take, for example, http://ibex.swri.edu/, a totally boss, relatively inexpensive program that Walt, TJ, and I helped launch recently. At our expected production levels we can produce ~10,000 Robobtboats for the price of that one mission (granted IBEX is doing some really far out science, but the point stands). Even so, since the sensors on satellites have such wide apertures and typically travel over great swaths of oceans and Earth, the cost per byte of valuable information may approach what Robotboats can achieve, in certain areas (maybe). Recently satellites have become excellent tools for analyzing the atmosphere beneath them, even going so far as to be able to identify particular elements dissolved in the air (btw, for many satellites, many of those capabilities go out the window when a simple cloud interlopes). However, there exist no technologies that can see underneath the surface of the water from space (that we know of anyway). That is why surface vessels are such a great tool. Our boats can look up to the sky, around the surface, and down into the sea simultaneously. You just can't do that anywhere but at the surface (or rather slightly above and slightly below the surface). In aerospace engineering there is a concept known as 'ground truth.' Basically the idea is that you'll need a few really high precision measurements to calibrate any data you pull down from a satellite. Robotboats are the best way we can think of to get those critical datapoints. Even more importantly, being able to correlate simultaneous co-local measurements above and below the surface of the water is crucial for high-fidelity scientific models.
It most certainly will. Robotboat don’t care. It will simply right itself and continue when it’s ready.
Yes! As we mentioned briefly in the description, we've done a LOT of testing. Most of it pretty boring, with the occasional epoch of outright terror. Our command and control protocol is pretty sophisticated, truly and entirely packetized, and based on military-grade satellite systems. We've successfully operated on several physical layers, including local RF, GSM/CELL, zigbee, and wifi. The Mk VI will also be outfitted with a satellite modem. We are confident that the telemetry system is essentially platform independent and capable of handling connectivity outtages.
All of the metal components are anodized or otherwise protected and the other surfaces will be coated in marine anti-fouling paint. We've had to think about stuff like how best to orient the solar panels to avoid bird droppings and salt deposits. We're investigating UV, advanced coatings, and mechanical wipers for critical sensor areas. Fouling may become a big problem as the boats are on station for months on end. That is one of the major reasons that is it critical that the boats be FAST, so that they can get out to station, execute their missions, and return home for refurbishment before fouling becomes an impediment to navigation and/or operations.
Will Robotboats pose a hazard to ships? How will you avoid other potential collisions and debris on the water?
One of the reasons that Robotboats are small is so that they will not pose a credible threat to existing vessels. Strong as they are, Robotboats will harmlessly bounce off of the hulls of almost all watercraft that carry people and goods. To people on pleasurecraft, jetskis, or tenders, Robotboats will be at least as visible as other traffic. At night Robotboats will be lit with red, green, and white in accordance with maritime standards. We do anticipate hazards in snagging on abandoned fishing nets, kelp etc. That is a very tricky problem. Much of the work of which we are most proud lies in our object recognition and avoidance scheme. When we say truly autonomous, we mean it.
The Mk I & II used off-the-shelf laptops (Mk I ran on windows 95!). For the the Mk III we went with a micro controller (Rabbit 3k series). As many of you may know from personal experience, debugging an embedded system is way harder than debugging 'normal' code (which may not be so easy itself!). By the time the Mk IV came along, we decided to modularize everything, which made swapping different control hardware a breeze (we also cut the number of control surfaces in half, so the control problem got a lot simpler!). The Mk IV & V have been run with a laptop, a low-power headless PC, couple different flavors of uC, a gumstix, and even an android phone. Slowly, we've been replacing sub-components with custom boards and FPGAs. The Mk VI computer is awesome... really low power. Rob is also using much of the same hardware he uses in the micro-satellites he builds, which means our system is as fault tolerant as they come (you can't go repair satellites, and we don't want to have to go repair Robotboats).
We haven't put a lot of effort into doing so, but the answer is yes. Many of the subsystems we've developed, especially the custom marine actuators, are probably quite useful in other areas. Wing sails make sailing a breeze (so punny around here). That being said, part of the allure of sailing is being at the helm amidst the sounds of a cloth sail luffing while ducking the boom. We don't expect that the recreational sailing market really wants this tech. We've been thinking about lots of other applications though, and http://news.cnet.com/8301-17938_105-57420784-1/cargo-ship-with-metal-sails-would-save-30-percent-fuel/.
These things ARE a sonar array. With a pod of as few as three networked boats operating in a small area, we can integrate the signals from hydrophones and use that data, along with the precise relative locations to back-out the origins of underwater sounds. Robotboats can also act as a relay for subsurface equipment via acoustic modem.
Sadly, there have been some accidents in the past that have resulted in tragic deaths of marine life. This http://blog.seattlepi.com/candacewhiting/2012/03/08/how-navy-sonar-kills-whales-and-dolphins/ is instructive. Keep in mind that the sonar systems with the deleterious side-effect that you may have heard of are many orders of magnitude more powerful than any system Robotboats could deploy. One of the applications we are most excited about is identifying the locations of susceptible creatures and ensuring the vacancy of any areas that could be affected prior to tests of such powerful systems. Robotboats can help establish go/no-go criteria in near realtime.
There have been at least two related projects on KS and there are several companies currently manufacturing related products. Personally, I think there is an abundance of opportunity for automated watercraft, & we would love to see a panoply of boats coming on-line in the next few years. We do think that Robotboat is a substantial improvement over everything that we have seen so far. The Scout project seems to have left an awful lot of energy on the table by not employing sails, and Protei, for all the wonderful public interest, media, and presentations they have generated, have yet to demonstrate a working prototype (that we can find). Plus we have some serious concerns about the feasibility of using a small craft to drag a heavy load. The X-1 from HarborWing is AWESOME, and its creator is probably the world's foremost expert on autonomous catamarans (coincidentally, a classmate of Rob's). However, we expect to be able to build ~1,000 Robotboats for the cost of one X-1. Liquid Robotics's waveglider is also a particularly elegant piece of technology, and there are many wavegliders already out there. We're really jazzed by their design - a mechanical AC 2 DC converter, smooth & strong. However, apparently wavegliders are very slow. See http://www.wired.com/gadgetlab/2012/05/wave-glider-crosses-pacific/ for more discussion as to why that is important. Also, if we are interpreting their technology correctly, the wavegliders are not able to operate in very shallow waters. We think there is an awful lot of http://www.pca.state.mn.us/index.php/water/water-types-and-programs/surface-water/lakes/lakes-and-lake-monitoring-in-minnesota.html to be done close to shore, and on shallow rivers and lakes, in addition to the open ocean. In the end, Robotboat only represents our best efforts, and, if someone has a better solution, we would love to see it. The world really does need this stuff!
This has always been a tough issue for us. We have chosen to keep many of the technical details of the project under wraps for now, but this is a somewhat uneasy compromise. We use open source tools all the time. Both Rob and I have been running at least one linux box since time immemorial. We are well aware that it is no exaggeration to say that this project (indeed most of the computing landscape as we know it) depends upon the work that others have made open. Also, we do recognize that going open source can be majorly beneficial in helping a project mature quickly. We do plan on making all of our designs, notes, whitepapers etc. free and open in the future, but we'd also like to a build a successful company around this technology. To date we have invested THOUSANDS of HOURS (not to mention hundreds of thousands of dollars) into this, and unlike university folks, we have no financial backing besides our day jobs. We discovered early on that the difficulties we face in developing the Robotboats are the EXACT reasons we need them; it is very difficult and expensive to operate on the water. We think it neither selfish nor unreasonable to try to get some of that investment back, and until that happens we will likely keep the details of our key technical innovations pretty close to the chest. As a compromise, we've decide to make the DATA ITSELF free (which is really the important part anyway).
Support this project
- (30 days)