In what is pretty stunning news to me, iRobot has purchased Nekton Research for $10 million USD. The reason this is so stunning, is because Nekton is 30min down the road from me, and I’ve interviewed there! Nekton is a small business located in Durham North Carolina and specializes in underwater robots. You can see one of their robots, the Transphibian, operating after the jump. Up till this point Nekton has based it’s business on government research contracts and hasn’t done any large scale product sales - this looks to change. iRobot says they plan on moving their “Maritime Programs office” to the Raleigh-Durham area. Eventually iRobot plans to have all their underwater research and development happen at this location. This will be where the Seaglider program eventually moves to.

iRobot has updated their website to reflect the new acquisition and features the Seaglider, Ranger, Transphibian, and “High Speed UUV“. I’m looking forward to see what exciting things will pop out of the water, so to speak.

(more…)

A 5.5m long, 6,100 lb robotic submarine called the Autosub6000 (I wonder if there was a 5000) is getting ready to explore the depths of the Cayman Trough in the Caribbean. The Cayman Trough, which you may remember from the film, The Abyss, has never been explored do to its incredible depths of up to 25,000 ft (the Mariana Trench, by comparison, is 29,000 ft deep).  The AUV will be used by the National Oceanography Centre in Southampton (NOCS) on two, month-long, autonomous, exploration trips to plumb the depths of the trough. Autosub6000 will be used to autonomously identify volcanic acvitivty, and mark the location for a ROV, Isis, to take samples. The unique features of the vents are expexted to yield many new species of life.

“They could be hotter than 500C (930F), and if they are that hot, they will probably have quite different chemistry and life forms - we expect to find new species.”

Good luck Autosub6000, stear clear of the aliens.

[Via Slashdot]

Pictured above is a group photo of all the teams competing in the 11th annual International Autonomous Underwater Vehicle competition. University of Maryland won the competition, unseating University of Florida from their 3 year winning streak. Congratulations to all the teams. The final standings for the top 8 teams are below. Pictured below is Maryland’s vehicle and a team photo.

Final Standings:

1. University of Maryland
2. University of Texas at Dallas
3. École de technologie supérieure
4. University of Florida
5. United States Naval Academy
6. University of Victoria
7. Cornell
8. Florida Atlantic University

Don’t forget that the Association for Unmanned Vehicle Systems International’s 11th Annual International Autonomous Underwater Vehicle Competition is taking place this weekend in San Diego, CA.

This year’s theme is “Underseas 11″ and will feature gambling themed tasks, such as grabbing a sack of cash from a “bank vault”, traversing an “air duct” (a floating PCV structure), and the like.

The competition is open to the public this Saturday and Sunday and is held at the U.S. Navy’s Space and Naval Warfare (SPAWAR) Systems Center’s Transducer Evaluation Center (TRANSDEC) pool located on Point Loma, San Diego. You can read the spectator FAQ (PDF) for more details.

25 teams from around the glob are registered for the competition, including my alma mater, NC State University. The teams have to construct a fully autonomous underwater vehicle that is capable of navigating, using machine vision, and acoustic navigation. It should be interesting (the competition is rather boring, since it’s underwater, but talking to the teams is fun). So, if you’re in the area, please go out and cheer them on. Send me some pictures if you do.

Update: The team from Cornell appears to be live-blogging the event.

Update 2: You can watch a live webcast of the finals here. It looks like the underwater video is being shot with an LBV from Seabotix.

Update 3: Here’s a list of the teams going into the finals.

Update 4: You can see the final standings here. Surprise ending!

The US Department of Defense (DoD) has released a report detailing the future of robotic military equipment and how the US military should proceed. The 188 page report (pdf) covers the next 25 years of unmanned military systems in the air, and sea, and on land.



This report is unique in that it address both ground and sea vehicles in addition to the much more "popular" and oft-used arial systems. This reinforces the Department’s position that integration of air, land, and sea systems is a vital part of the future of US military systems.

The report details recomendations made by field commanders and how further AI developments can lead to "thinking" systems.


Another fascinating aspect of the report is a huge appendix listing all the unmanned systems currently in use by the military. The image to the left is a shot of the Lockheed Martin Protector Unmanned Surface Vehicle.

If you have a hard time downloading the document, Roland Piquepaille provides a good outline.


(more…)

Ah, back home to North Carolina where the tea is sweet (very sweet) and the humidity makes it feel 15 degrees hotter.

 As promised, I’m here to provide closure to NCSU’s journey to San Diego. And, no, we didn’t place in the competition. We were so very close - one of the top contenders actually. We were able to demonstrate in practice all of the tasks, which, as far as I know, no other teams were able to do.

 So let me just give you blow by blow starting Saturday morning:

 

First Preliminary Run

Our robot has a network cable teather which allows us to recompile code and run it on-the-fly. We consulted with David Novak, the in-charge-man for the competition rules, and he assured us we could run with the teather as long as we stepped away from the computer once we started our run.

However, when we arrived at the platform, the chief judge told us "nothin’ doin’" and made us remove the cable and put in a blank puck (aluminum endcap where the cable origianlly entered the electronics tube).

This was a high stress situation and everyone stood up to it bravely, but unfortunately, the robot failed to run.

A heated altercation then followed. We tried, as courteously as possible, to get some sort of concession since we had been misinformed. Eventually they agreed to let us run again that evening at 6:00.
 
Much stress ensued …

Second (1.1?) Preliminary Run

At least we had another chance on Saturday. We buttoned up the robot. Brought it to the launch platform, lowered it into the water, and nothing happened. OpenCV kept crashing, so our visual starting device (an orange "key") wouldn’t work.

In a mad, heart-stopping, dash, Matt opened up the robot and reconnected the network cable for Sterling to put in a delayed start mechanism. In a flash, the robot was back together and in the water.

We bagan to count the seconds … slowly, slowly, our 5 minute timer ticked down and we waited for the robot to start.

Nothing happened. 

Heat-broken, we pulled the robot out once again, and prepared for a very long night.

 
Final Preliminary Run

We had stayed up all night working on the robot. Trouble shooting bugs and trying to install a wireless network card so we could wireless start Seawolf.

Finally, the 11th hour dawned and to our horror, we found out our thrusters wouldn’t work. Nothing. Zip. Matt rushed back to the hotel to get the spares. The spares didn’t help. Some thrusters would cut on. We’d power cycle and a different thruster would cut on. It was a nightmare. No rhyme. No reason. Utter confusion.

Eventually, we found a combination that worked. It was very touchy, but it worked. We tried it 4 times at our booth. Twice at the starting platform. And finally, once on the dock.

However, when we lowered the robot into the water. It once again refused to work. Seawolf got stage fright.

To add to our misery, at this same time our batteries failed. So, we hauled Seawolf out of the water and did a NASCAR breakdown and battery replacement of the robot. There were cameras and people all around us taking pictures in amazement. It was pretty spectacular. The whole process took about 3 minutes.

With just a few minutes to spare we once again tried to coax Seawolf into playing, but he refused, and simply cut on two thrusters

and

spiraled

down

into

the

depths …

Conclusion


It was all worth it. Sure, we’re horribly disappointed, but it was an amazing experience. Everyone loved the robot - the media (on live San Diego TV), visitors ("I saw your robot on TV!"), and the divers ("f***-it, I like Seawolf better!").



Great job University of Florida you had an awesome robot and deserved the title. But, enjoy it while it lasts, cause next year

NCSU will be king!

Well, it’s 10:21 AM. Our first preliminary run starts at 11:00 AM. As of now, two teams have made it through the gate, and two teams haven’t showed up. One team, I think the University of Florida, navigated to the acoustic pinger and surfaced.

We also have another run tomorrow morning at 10:00 AM.

The results of these runs will determine whether we make it into the finals. We can navigate through the gate just fine. We spent all night hashing out the acoustic navigation software and we’re very confident that we can find the pinger and hopefully surface. If the robot doesn’t surface we set a timeout, because we know it finds the pinger but it has trouble surfacing.

After the pinger we’re going for the pipeline then the docking station. We chose to do the docking station last because 1) it has the least amount of points associated with it and 2) there’s a chance we’ll start following the sun and run in to the wall - which would be a disqualification.


Mike, Erin, and Jim


Transdec

It’s almost 9PST and I’m standing here blogging pool (amazingly large dive tank?)-side. Our arrival time was pushed up one hour this morning to get teams here in time for some live coverage from a local San Diego TV station.
…
Yeh, definitely got up at 5 AM.

Mike and I walked into the command/al queda room to find no robot. Mike said, "surely it isn’t in the pool now … " Sure enough, Frankie/Jim/Sterling were at the pool, and had been all night testing.

We buttoned up the robot and got onsite around 6 AM. The TV guy wanted to show our robot on TV so we hoisted it up in the air and were on live TV. We’re trying to get a copy of the tape. Your’s truely was the PR guy and did an awesome job pimping the robot to an uneducated CA audience.

HOWEVER, After about 2 hrs, this guy is getting aggravating. He keeps wanting to show the robot during commercial breaks and it’s interrupting our testing.

We’ve got a TON of work to do and need as much uninterrupted time as possible.

Oh, and as of 9AM  MIT and Cornell haven’t showed up today. They’ve won first places in previous years, so if they’re having trouble this could be a big break for the rest of us.

I’ve got a lot of audio but it’s taking a long time to edit. Fingers crossed I’ll have the first podcast up this afternoon. In the mean time, here’s a few pictures from the first day of practice.


shot of our setup


Here’s Seawolf being hoisted into the water for the first time.


Behold the beauty! Mike spent an hour and a half putting those stickers on. He’s OCD about it.

Well, I lied. Apparently I do have my camera cable. So, here’s some pictures from earlier today.

When the sun goes down, it gets pretty chilly here. Matt was freezing his butt off in the pool for a couple of hours, and now Jim is in the pool.


NCSU


University of Florida


Montreal

Day one of the NCSU Underwater Robotics team’s trip to San Diego for the International Autonomous Underwater Vehicle Competition has come and gone. We arrived in San Diego around 10 AM after some miserable flight time. We waited around in the airport for about 2 hours for our luggage and a proper sized vehicle.

We set up one of our hotel rooms "al queda" style with all the equipment and the robot. Frankie (team captain) and Sterling (software god) are staying in that room and started setting everything up while myself (controls guru, etc.), Mike (wiring master), and Jim (genius Indian) went off to buy stuff.

San Diego drivers are crazy and we spent a very long time looking for a Target/Walmart and  Lowes. Six or so hours later we finally arrived back to mission headquarters.

One of the problems we’ve been having is that our dropper box is interfereing with our DVL (Doppler Velocity Log, aka extremely expensive magic box). Thanks to the genius of Ken Boone we bought ourselves a can of pineapples, ate the pineapples, and ghetto rigged the can around our dropper box to magnetically shield it from the DVL.

We spent the evening half dead from lack of sleeping while working on trying to get our electronics tube back together.

Tomorrow we go to the introduction meeting for the contest.
I’ll be posting episode one of the podcast sometime today, so stay tuned for that.