Wired’s Eric Smalley has an awesomely titled article about a Massachusetts based startup, Harvest Automation, is testing a small farming robot to work in nurseries in the horticulture industry.

The Harvest Automation robots are knee-high, wheeled machines. Each robot has a gripper for grasping pots, a deck for carrying pots, and an array of sensors to keep track of where it is and what’s around it. Teams of robots zip around nursery fields, single-mindedly spacing and grouping plants. Think Wall-E without the doe eyes and cuddly personality, or the little forest-tending ‘bots in the 1972 sci-fi classic Silent Running.

Thank you Wired!

Just watch this video – Words, I have none.

Working with similar robots, I can only hope that one day I can make them move that smooth and agile.

Zenta’s latest project is a morphing robot. That’s right, a MORPHING ROBOT.

I tossed this brilliant Norwegian a few questions, and he was nice enough to get back to me!

What’s your background and how did you get into robotics?

My background is engineering. I’ve always been fascinated by all kind of robots since I was about 6 or 7 years. (I’m now 38 year) .I began building robots very early, using mostly meccano. At some point I found meccano a bit limiting and started making custom parts. But I didn’t get more serious into hobby robots before 2006 when I started on my rather famous Phoenix hexapod.

The main reason for why I wanted to construct a biped is that I had some ideas of how to control it using Inverse Kinematics (IK) based on the current code used for controlling hexapods. The main difference is balancing/stability. Also gear-slope/backlash in the gear of the servos is a challenge on bipeds. Keeping the weight as low as possible is also more important on bipeds. When it comes to stability I tried solving that on Archer by simply moving electronics and battery from side to side.

Thanks, glad you liked it! I got the idea for building MorpHex for over a year ago. Mainy I got the idea watching my two eldest kids playing with Bakugan and I thought it would be cool to make a hexapod that could transform into a sphere and back to a hex again. I’m still working on MorpHex, you can see what I’ve accomplished so far on my blog. The main challenge with MorpHex was to make the variable sized body, the body needed to expand for making more space between the leg sections. There has been several difficulties during the making of MorpHex. At some point I considered to end the project, so I had to do some changes to the leg design for making the robot less heavy and also for removing some conflict between the sphere sections.

The smooth movement are simply accomplished by accurate control, correct math and correct timing. Jeroen Janssen’s work on the Phoenix hexapod Basic Atom Pro code make this rather easy to accomplish. Instead of using a PS2 controller, I’m using a custom made remote controller. The remote controller play a big role for accurate control.

Oh, thats very hard to say. I enjoyed them all. I think I’ve to say Phoenix was the project I enjoyed most.

A full humanoid biped, a new hexapod and or a new quad…

Lego mindstorm or VEX Robotic System is a very good plattfom. I’ve to say I’ve not worked with either of them. But I plan to introduce Lego Mindstorm for my kids one day. Also, a hex or biped Brat kit from Lynxmotion is a good start.

HOLY MOLY(bdenum), Robo Bat Friends! Let me tell you about this new iPhone iPad iPod app for $2.99 that-
And I quote:

“Description

No 1 Staff Favorite in Japan!
LOVEBOTS by KIT ROBOT has arrived! The NEW Digital Robot Assembly Kit and Robot Avatar Creator, SUITABLE FOR ALL AGES.

Placing creative power in your hands! Join the loveBot
Presented by Award Winning WhiteWall Studios.

Forget the Golden Plastic Age of model kits, limiting the range of parts to one model and WELCOME to the Golden Digital Age.

Now Introducing LOVEBOTS, where you can create over 1.3 MILLION combinations of Robots out of the box!”

Check out the gallery of ideas on boingboing as there are lots of delicious and eye popping pictures to ponder while you get your iTunes on speed dial.

I leave you now with one final quote:
“We are asking LoveBot fans to send us their creations, as we plan to release The Art of LoveBots app book. “

I don’t know about you, but I have been yearning for a more coherent way to shop online. It looks like the Estonian startup Fits.me is onto something L here. Or S or XXL or OMG.

I’m very excited about this, even though, at present, the Robot Mannequin only shows sizing for men. They recently closed a second round of funding, though, so it looks very promising for women, children, robots, animals, mutants, cartoons, RPG’rs, toys, furniture, fonts, ok ok. The mind reels.

Special thanks to new robot pal, JarGon, that lovable villain, from getsatisfaction.com for the tip!

Artist, Designer, Fabricator, Collaborator, Customizer, Collector Whew! He’s busier than a one legged man in an ass-kicking contest who designs, builds and sends original robot toys all over the world to collectors and galleries. Though the figures have no traditional “working parts”, they do have incredibly believable details and lovely backstories. Cris is particularly interested in the relationship between robots and nature, and each character shows its wear and tear with pride. I hope to catch up with Cris soon and find out more about his robots, their histories and the art of decay.

This week, we bring you the lovely Nick Donaldson, of GotRobots.com. Nick is often referred to as “That one guy. From RoboGames. With that monkey. The one that sits on his shoulder and high-fives people.” In any case, Nick is a brilliant inventor and robot builder and we were thrilled to have a chance to talk to him!

Intern: Thanks for taking the time to be interviewed by SuicideBots! Who the heck are you?

Nick: Nick Donaldson, roboteer extraordinaire!

Intern: How’d you get your start in robotics? What makes you a “roboteer”?

Nick: I have a background in mechanical engineering and started my career building laser optomechanical windtunnel probes. I then spent about a decade writing software for the burgeoning internet starting in the mid-90s.Then came Mindstorms! My wife Rebecca gave me one of the first Lego MS sets when they were launched. It got me started combining my love of mechanical and software into robots. I started out with a few simple Lego robots and then soon beanched out into walking robots. I found however that as they got larger and more complex the plastic flexed too much so I invested in the machine tools I needed to start building them out of aluminium, and so was born Ziggy, my first real robot, built in 2001 and still going strong. [6 times gold medal winner at RoboGames]

Nick Donaldson’s “Ziggy”

Intern: Tell me about building laser “optomechanical windtunnel probes” — what are they? How do they work?

Nick: There were two types, based on a similar technology – laser velocimeters took a single coherent laser beam, separated it into two, then refocused the two beams down to intersect in a very small volume where they interfered with each other to form light and dark fringes. As a smoke particle passed through this volume it would flash as it passed through the fringes. By measuring the frequency of flashing you could calculate velocity orthogonal to the fringes. Thus, by using three different wavelengths of light at right angles you could get all 3 vector magnitudes and calculate the speed and direction of a particle.

Intern: So basically, it’s a super-accurate measurement of aerodynamics?

Nick: Exactly. Further processing using fourier transforms would allow for sizing of the particle also, so you’d end up with a volume of a few microns in 3d-space that you could traverse around inside a wind tunnel (through a window), and so build up a detailed picture of the flow around a wing or car or whatever. Through this work, I got to design and install systems at NASA Ames in Mountain View. They have some seriously HUGE wind tunnels – big enough for a full size jumbo jet! Very cool stuff.

Intern: What were some of your very first robots like, as compared to Flik, your walking LEGO bot, and Ziggy, your walking aluminum bot?

Flik, a walking robot made of LEGO pieces

Nick: My first Lego robot was called Tobermory and was a fairly simple tracked bot with bump sensor avoidance, but extra big eyes with eyelids that would react to a bright light by squealing, closing his eyelids and backing away – kinda cute. I also built an 8 legged walker that used a rotating cam system.

Intern: How have your designs, and thoughts on robot design changed as you’ve continued to grow as a roboteer extraordinaire?

Nick: There are certain tricks that you figure out and then re-use, so as your experience grows you have more of a toolkit as it were to integrate into your designs. For instance, with the walkers I realized that if I designed the linkage mechanisms to straighten out when the foot is on the ground, no motor power is needed to keep the leg down so you rbatteries last longer. Since I figured that out, every walker I have built uses it. I even redesigned Flik (the LEGO hexapod) to use it.

Intern: Very cool! So what are you working on these days?

Nick: I have been working to commercialise the shoulder sitting monkey (Cheeky Monkey) as a toy.

Intern: What’s the Cheeky Monkey?

Nick: Cheeky Monkey is your shoulder-sitting buddy, he’s a small animatronic monkey that attaches to your shoulder and has a life of his own. He looks around, waves, scratches his ear, and you can also send him commands to high five your friends or wave at people! I originally built him for fun but had such a huge reaction any time I took him out in public that I decided to redesign the mechanism using just two tiny motors plus a bunch of cams. We are now in the final stages of bringing it out as a toy. It will be coming out this summer for about $25. I have been working closely with manufacturing in China to get the final details right.

Intern: Very cool! What else have you been working on recently?

Nick: Well, I am also designing robots professionally – over the summer last year I designed a very animated robotic head for Meka Robotics in San Francisco.

Nick: I am now working on a Nurse’s assistant; a robot to lift patients out of bed and onto a gurney.

Intern: Wow- that sounds like it would require a lot of engineering to work reliably, painlessly, and unobtrusively. How much can you share about the strategy of making that machine work?

Nick: It is surprisingly hard to build a machine strong enough to lift a human – we’re heavy! Some very expensive technology is needed; we’re using harmonic drives as the heavy lifting workhorses, since they’re perfectly suited to high-end robotics.But very expensive. We are also using Series Elastic Actuators to allow the robot to feel it’s environment – it can react to force, not just to positional information, which is important in a dynamic environment where a human may wander into the path of the robot.

Intern: That’s awesome! Thanks again for the interview! Any last words?

Nick: My army of robotic monkeys will take over the world! Mwah-hah-hah-hah!!

Ray Billings’ Last Rites vs VD6

If ever there was a ComBot that made you say “OH $#!%- RUUUUNNNN!!!!” it was most likely built by Ray Billings, of HardCore Robotics. Ray is an ex-prison guard, and became interested in combat robotics after the fact, when he was working at an adult college, and a colleague introduced him to the sport. He was hooked, and so this week, our BotLight is on Ray Billings. Interview below!

Intern: What was your first venture into combat robotics? How did it go? What did you learn?

Ray: Our fist event was Battlebots 4. We took one middleweight robot, a thwackbot style machine called Darkness. We ended up gong 3-1 and making it to TV rounds, but never actually made it to TV. The frame for this first machine was made from an old bedframe. We had a lot of fun, and certainly learned a lot about competing in general- I was astonished at the overall activity level in the pits, something I wasn’t prepared for.

Intern: What kind of activity?

Ray: For one thing, I was amazed (and still am today actually) of how many machines came to the event without being combat ready. When I show up to an event, I can literally pull my bot from the truck and drop it in the arena to fight instantly, which is one of the MAJOR differences between the great teams,and the not so great ones in my opinion, anyway.

The pits at ComBots are always alive with activity!

Another thing that surprised me was the frantic repairs needed between matches. I had done some various forms of racing in the past (carts, motorcycles, etc) and have had to deal with that kind of activity before, but bots take it up a notch, no doubt about that.

Intern: How did your ideas and designs develop during and after your first competition?

Ray: I have always built very tough machines physically, but the first few machines were not all that impressive as far as delivering damage to the other guy. Battlebots had a rather odd (and quite limiting) arrangement, requiring you to be ranked for getting to the TV rounds,and requiring you to keep the robot in the same configuration. For BB5 I would have loved to have built a different machine, but since I was on the verge of being ranked with Darkness, I was forced to continue with the thwackbot design. For BB5 I rebuilt Darkness from scratch, and again went 3-1 with it, which would have put me in the ranked category for the class. Unfortunately, BB never had season 6, so my ranking meant nothing after that. However, this meant I was free to build what I wanted, and so I started building the big kinetic energy (KE) bots I am known for today.

Intern: Very cool. On that note, you and your team have become well-known for having some of the biggest, baddest ‘bots around. How do you determine how much weight to put towards a weapon, vs how much to use on defense, vs how much drive? Do you consider your KE ‘bots “glass cannons?”

Ray: Glass cannons – I like that term! There have been a LOT of one hit wonders, machines with high KE but no real chance to use it in a sustained fashion. For me, I try to get the absolute most energy possible in a machine capable of delivering repeatedly, and reliably. Sometimes I succeed, but often times I don’t. If I’m in a long tournament, the wear and tear will eventually get to the point where I can’t keep the ‘bots running. I’m always on that fine line between maximum KE and being able to survive.

I usally start by looking at the weapon – between 1/4 and 1/3 of the mass of the bot will be the actual spinning part. More than this and I won’t be able to build a strong enough frame to make it survive, and less than this will mean I won’t have the lethal power I am after

Next, I look to a drive system that is capable of delivering the weapon. No pushing, no real speed, just barely enough to reliably point the bot in the right direction. I build weapons, not pushers, so drive only needs to be able to get it there.

After that is batteries – my bots are harder on batteries than anyone in the sport, and so I need to plan carefully. I need enough to be able to source the amp current for the big weapon, and enough capacity for at least 2 minutes. Matches are 3 minutes long, but I plan for 2 minutes. If the other guy can take 2 minutes of me smacking him, he deserves to win.

Whatever weight is left over goes to frame and armor, and frame is WAY more important in my designs than armor.The structure supporting the weapon needs to be substantial – I usually have to build a new frame after every event, or every other event, because it gets bent and twisted from the forces being unleashed.

Intern: Wow! What can you do to make sure the frame is solid enough to hold up to those forces?

Ray: I use 4130 chrome molly tubing for the frames, and almost exclusively use 1″ diameter tubing. However, the wall thicknesses vary consderably, depending on what I think the stresses are at that point. Some of it is very thin, at .035 wall, whereas I have used some as thick as .250 wall. It looks like it is all the same tubing, but it isn’t. Sort of a trade secret for building a strong but light frame. Since I rebuild so many frames, I have gotten pretty good a gauging stresses and building accordingly. The probem is I keep building for more power, so the stresses keep increasing .

Intern: Have you ever considered using computer software to aid you in gauging the stresses? What are your thoughts in that respect?

Ray: I use AutoCAD and Solidworks on occasion, but more from a design process of “will it fit”. It’s my opinion that the tools in computer modeling just can’t take into account the unpredictable forces that come in a combat environment. More power to those that want to use them, and I am a big fan of using all the tools at your disposal, but I think you would get answers that would fail in the arena.

Intern: What’s next for Hardcore Robotics? Any design changes or new bots on the table?

Ray: Our middleweight, Mortician came home from the ComBots event without any major damage. The few things on it needing work are done, and it is combat ready now for RoboGames. Pretty much the same story for our 30 pound bot Mitch. The Heayweight Last Rites is pretty much unfixable, and will need to be rebuilt from scratch.

We are also in the process of developing a new 30 pound machine, which will be something we have never tried before. It will be a high KE spinner, but in a different configuration than we have attempted in the past. If development proves successful and it plays out well at Robogames, we may try to scale it up for the MW/HW classes. And if time allows, we may even bring an easter themed bot for crowd appeal!

Intern: Very cool! Thanks again for taking the time get interviewed! Do you have any parting advice for people interested in getting involved in events like ComBots, who haven’t yet made the jump into the sport?

Ray: There are a lot of folks that I have talked to who would say something like “I’d like to do that, but I just don’t think I would be able to”. It really is amazing how easy it can be once you get started actually. There are some up front costs fr electronics,but after that you can make competitive and fun bots out of scrap material – like my bedframe. The builders who compete are all really good about advice, and if you have questions just ask someone. They’ll talk your ears off !

Team HardCore Robotics (Ray at center)

Photo Credit: Dave Schumaker

In November 2001, Matt & Wendy went and saw fighting robots live for the first time after having seen them on television. Bitten by the robot bug, they began designs for the ‘bot that would ultimately become the crowd pleasing Sewer Snake!

Exclusive image of Sewer Snake’s earliest design

Initially, Sewer Snake’s design had a spinning bar weapon, much like his arch enemy Last Rites. That design was scrapped for a high-RPM drum weapon like Touro [Say- we interviewed those guys, too!]. When everything but the drum weapon on the ‘bot was said and done, the PlumbCrazysons (Wendy and Matt) decided to compete in an event that didn’t allow spinning weapons, which meant scrapping the drum weapon for something that at least looked incapable of killing or injuring passersby. Thus, Sewer Snake and his flipper were born.

Classic Sewer Snake

In that first competition in November 2002, Team PlumbCrazy won their first round, only to have a gearbox explode [spectacularly, we hope!] in their second fight, causing them to be eliminated from the competition. Wendy & Matt were hooked. In 2004, they competed in 7 competitions, amongst various other demonstrations. With 3 bots in 3 different weight classes, they had more than a handful.

Interestingly, Wendy says that she doesn’t like the idea of going in as top-dog in competition. She argues that it’s fun to go in as the underdog, without some big, scary weapon that looks like it could kill you. It’s a lot more fun, she says, because it’s exciting to have to really work for a win. Driving skill becomes a much bigger factor when you have to dodge bullets and find weak points, as opposed to mowing through anything in your way.

Robotic combat is just like any other sport – it’s constantly evolving, and every year the competition gets more fierce. The armor gets thicker and the weapons get stronger, and every year the builders come back for more. In 2005, Team PlumbCrazy designed Angry Asp, a middleweight bot with a spinning drum weapon.

Angry Asp took a bronze medal at the 2005 RoboGames, but has since been modified to have spikes and a flamethrower, in the interest of making the competition more exciting.

Above all else, PlumbCrazians Matt and Wendy look for presentation when building and improving their bots. What will please the crowd the most? (Hint: Big, bright tires!). Wendy believes that often, the biggest factor of winning in ComBots is driving skill. Sure, you can put a thousand hours and a million bucks into a solid titanium robot OF DOOM, but if your driver can’t match the awesomeness of your robot OF DOOM, well, you may as well stay home.

If you go to RoboGames 2010, Wendy & Matt will be hard to miss. Be sure to thank ‘em for donating their time to be interviewed by SuicideBots, and for donating their tools that one time, to be abused by that one guy called Intern [Wait a minute...].

<– AKA that guy.

Marco has well over a decade of hands-on robotics experience, and has been involved in ComBots since 2003. He graciously offered his time and insight on ComBots, teamwork, and how to avoid pummeling random passersby with 120lb steel cylinders from 100 yards away! See our exclusive interview with the founder of the RioBotz robotics team below!

Intern: Thanks for taking the time to talk about RioBotz

Marco: You’re welcome.

Intern: If you would, let’s start with a quick history of RioBotz- who are you guys? Where are you from? What do you do?

Marco: I’m a professor from the Pontifical Catholic University from Rio de Janeiro, Brazil. I coordinate our team of about 20 undergraduate students building ComBots, sumo bots, and hockey bots.

Intern: How long have you been involved in robotics, and how did you start?

Marco: My first hands-on robotics training was at 1996, during my Ph.D. at MIT. But only manipulators and peaceful rovers, none of the combat stuff. I finished my Ph.D. in 2000, when I became a professor back in Brazil. In 2003 a group of 5 students talked to me about combat robotics, and I was hooked. We build our first ComBot in 2003, a middleweight named Lacraia. We’ve learned a lot since then, building ComBots is very different than building regular robots that don’t have to take such kind of punishment.

Intern: Very cool. What kind of research did you do initially when you decided to get into robot combat, before building Lacraia?

Marco: In the US, I used to work with manipulator control for nuclear power plants, and had a lot of contact with prototypes for Mars rovers. In Brazil, I was working back then building inspection mobile robots (also for nuclear power plants) and with manipulator control for subsea interventions for the oil industry. We learned a lot from building Lacraia, we’ve realized that while the control part is relatively simple, the mechanics and electronics are much more challenging. In 2004 we built another robot, Ciclone, a horizontal bar spinner (Lacraia was an overhead thwackbot).

Intern: What was Lacraia’s design, and how have your ideas about good ComBot design changed since then?

Marco: Lacraia was very weak for today’s standards. Its 1mm thick aluminum armor was a joke, but it was all weight we had left after using standard (meaning heavy) industrial parts. Then we’ve learned a lot about high-end motors, gearboxes, batteries, which helped to create Ciclone in 2004, later Titan (also bar spinner) in 2005, and Touro (drumbot) in 2006. This also helped me with my other research and industry projects, building tougher robots.

Lacraia, with its delicious, thin aluminum armor!

Intern: Nice. How has RioBotz as a team developed since 2003?

Marco: In 2003 we didn’t have any funding, so Lacraia was built with inexpensive parts. We got 6th place in the Brazilian nationals in 2003. Then the university started to help with some funding, the number of students grew from 5 to maybe 12, and we’ve built Ciclone. Ciclone was the Brazilian champion in 2004, and later in 2005. This helped us getting a sponsorship from an oil-related company, and then building Touro. Touro and our 60 pounder Touro Light got gold at RoboGames 2007, this was great for media exposure in Brazil. Currently we’re out of sponsors, but still looking. The university helps a lot, and I contribute quite a bit, myself for this passion. We’ve built ComBots in all combat categories, except super-heavyweight.

Intern: So now the big question: Who gets to drive your ‘bots?

Marco: Until 2006, we had a very good driver named Felipe. After he finished his Masters and left the university we found a new one, Daniel. Daniel’s debut driving ComBots was at RoboGames 2007, when he won 2 golds! I think he’s played too much videogames to be able to learn so fast.

Intern: On that note, how do your drivers prepare for events?

Marco: We’ve bought several large steel plates, that we tile on campus to be able to drive the bots without causing damage. Training is basically driving with the weapon off, driving with the weapon on (it’s very diffrent due to the gyro effect) and weapon tests. Most of these tests need to be done after midnight, when the university is empty. There’s not a lot of space on the steel plates, which helps to learn how to maneuver. For weapon tests, we use a 40lb aluminum block and a 120lb steel cyclinder as sparrings to take the punishment from Touro’s drum.

Intern: Have you ever been surprised by the power of one of the bots?

Marco: All the time. Ciclone’s first test was on the university’s parking lot, and nobody was prepared for that. There were monitor pieces all over. Last year, when we started using a single tooth drum in Touro, we were also surprised. In one of the hits, the 40lb aluminum block was thrown about 100 yards, it landed on the other side of a small river that divides the campus.

Touro – Some say RioBotz are nuts. We say they’re bolts.

Intern: Wow! What do you do to ensure that nobody gets hurt when testing?

Marco: This is always after midnight, and a few students go to all possible entrances to the test site, to make sure nobody passes by. We don’t have an arena, so the other students just move away and cover, using goggles. The main problem is that students start to get confident with the robot and start becoming wreckless, I have to shout all the time for them to get really away from the bot.

Intern: So that being said, what’s coming next for RioBotz? Will we be seeing you at the 2010 International RoboGames?

Marco: Yes, for sure. We’ll bring bots to compete in all combat categories, except super heavy and auton-ant. We’ll also bring our R/C and auton 3kg sumo bots, and it will be the international debut of our robot hockey team. We’re currently rebuilding all Touro bots from 12lb to 120lb (Touro Jr, Touro Feather, Touro Light and Touro). Touro’s chassis was basically the same as the one that competed at RoboGames 2006, so we had to upgrade it to face the tougher and tougher competition. They’ll still be drumbots, but now we’re trying a new design for the structure, to help them get stronger, easier to maintain during pitstops, and hopefully having a smaller chance of getting stuck in the arena.  I’m proud to announce that last week, Touro became the first ComBot to be commanded by thought! A masters student whom I advise has developed a very compact brain computer interface which can command Touro (or other bot) to move around just by reading electrodes from a cap he wears. He’s achieved great results, with only 1% errors. We’ll present it at the IEEE ICRA conference in Alaska in May, including nice videos from “Psychic Touro!”

Intern: Very cool! Do you think we’ll see Daniel driving around Touro in combat with electrodes in his cap one day?

Marco: It is possible. But probably it wouldn’t work out well, his hands can maneuver much faster than the brain interface. Besides, if some beautiful girl would pass by Daniel, Touro would get out of control with the resulting thoughts!:-)

Intern: We’re looking forward to seeing you at RoboGames 2010! Thanks again for offering your time and expertise- do you have any parting words of advice for future robot builders out there?

Marco: Yes, research a lot. Read all available books, search for websites with build reports, and study the design of successful bots. Try to always question “why was it done that way?” and try to adapt the positive points to your own design. You’ll save years of broken parts and bots if you spend a few months studying other bots.

Marco (center) among the RioBotz at RoboGames!

From the NYT:

The new vehicle is to replace Alvin, which was the first submersible to illuminate the rusting hulk of the Titanic and the first to carry scientists down to discover the bizarre ecosystems of tube worms and other strange creatures that thrive in icy darkness.

The United States used to have several submersibles — tiny submarines that dive extraordinarily deep. Alvin is the only one left, and after more than four decades of probing the sea’s depths it is to be retired. Its replacement, costing some $50 million, is to go deeper, move faster, stay down longer, cut the dark better, carry more scientific gear and maybe — just maybe — open a new era of exploration.

I personally will definitely be pouring one out for my junked homie; discovering the Titanic and all the attendant wonder of deep sea exploration made realize how bitchin’ science actually is. Alvin is where the scientists who deployed Argo the ROV and Jason the autonomous submersible were stationed.

Badass leader of robots, we salute you.

Enjoy your retirement.

[via Gizmodo and the Grey Lady]