New Video Of Sand Flea Robot Leaping (Onto) Tall Buildings 30ft High | Hacker News Reader

New Video Of Sand Flea Robot Leaping (Onto) Tall Buildings 30ft High

singularityhub.com

94 pointszacharye24 comments

I think the most interesting thing about this robot is how it maneuvers while in flight. If you watch closely, you will see that the robot stabilizes into the 'flat' configuration before reaching the apex of its flight. Without stabilization, it would simply tumble end-over-end.

I believe the robot uses its wheels as reaction wheels, adding the proper torques to the tumbling body to cancel out longitudinal angular velocities. There are not enough controls to cancel out rolling, if it launches wrong. The robot would need an internal lateral reaction wheel or CMG cluster for that.

This is a pretty cool robot and one I intend to copy at my earliest convenience! If they combine it with the Van der waal's [1] adhesion effect then you have a robot that should be a really good urban navigator.

Funny story, I designed a robot based on the hopping principle for the first DARPA challenge (the one where you were in the desert trying to go from Barstow to Las Vegas) except that we ran into issues with the ground being too soft in places to reliably jump. And the fact that we had an on board compressor which both weighed too much and took too long to recharge the jump tanks, but other than that :-)

[1] https://en.wikipedia.org/wiki/Van_der_Waals_force

> If they combine it with the Van der waal's adhesion effect then you have a robot that should be a really good urban navigator.

It's funny you mention that. I went to grad school with the guy who created the jumping robot. His PhD research was in creating artificial Gecko footpads for robots.

And when that version comes out we'll have a product where someone throws a robot out the window, it jumps up on to a wall of a nearby building, and starts streaming back video of the area. Or, in the less fun case, it will jump on top of a vehicle where the armor is weakest and fire a shaped charge straight down.

Nice. My friend did his PhD on gecko adhesives too[1]. He later built[2] a "six-legged" robot like Boston Dynamics' RHex 'bot[3].

[1] http://northdesignlabs.com/?page_id=12 [2] http://northdesignlabs.com/?p=38 [3] http://www.youtube.com/watch?v=ISznqY3kESI

Do you have anything online about that robot? I thought of building a hopper/glider hybrid a few years back. It would basically work like a large grasshopper. Windup and release a spring to jump and at the apex of the curve release a small wing to glide. Lather. Rinse. Repeat.

It was to be an experiment in solar powered robots that would use the spring as energy storage instead of a battery.

Never built it; this has me thinking about it again, tho.

The idea grew out of our Battlebot, one of our sponsors was Festo which makes pneumatic actuators. They had a new actuator which was a short throw 6" diameter cylinder and offered us one for our 'bot. The idea was that if we mounted it into the front of our wedge, and then fired it when an opposing robot was on the front, it would knock them off. Doing the calculations (150 PSI, 4L expansion tank) showed that we could develop an initial thrust of 2,100lbs with a working distance of 2". That calculation had the opposing 220 lb robot experiencing an acceleration of something like 1000 ft/sec^2 for .004 seconds. The end result was that we were in danger of tossing the opponent out of the 'battle box' by launching it over the top (in practice this wouldn't happen because the chance of firing the weapon under the exact center of mass was infinitesimally small) but it got us thinking about the hopper idea for the DARPA challenge (we already knew our robot could withstand the force of falling onto concrete from a height of 20').

Providing the gas pressure however is an issue. You can use solid CO2 recharge (drop in some dry ice, seal, and warm) but for the challenge it had to go a looooong way. And that would require a fuel, a generator, and a compressor. Weight went up, survivability went down, and some early tests on what was essentially a pogo stick pointed that it wasn't practical at those mass levels.

Gliding would have been an interesting choice. With something light like a parafoil wing even more so. If I dig up my notes I'll put up something on my web site about it.

I should caveat subsequent comments by saying that I would have given my childhood for this toy (and therein I would have learnt the meaning of irony too, but I digress).

Anyway I am wondering what this could really be intended for. If it is indeed meant to carry a camera payload for cheap soldier surveillance, then I have an alternative suggestion: http://www.amazon.com/Parrot-AR-Drone-Quadricopter-Controlle.... This $298 quad-copter with two mounted cameras seems like it could do the job just fine. (Incidentally on the subject of cheap RC toys saving soldiers' lives and proving themselves viable short-term options: http://www.rccaraction.com/blog/2011/08/04/traxxas-rc-truck-...). Anyway, especially with Petman (http://www.bostondynamics.com/robot_petman.html) purportedly being for chemical suit testing (which in my opinion is like saying crash dummies need to have guns and machine vision), Boston Dynamics has shown that it likes releasing potentially mid-path projects for other-than-ultimately-intended purposes (makes perfect sense from a testing perspective). So I wonder what Sand Flea could be working towards other than surveillance? 11 lbs./30 ft is not bad considering a quad copter of similar size would be probably max 4-6 lbs. So maybe they are gearing it up for heavier payloads. Nothing like the 400 lbs alpha dog carries (at that point, assuming similar tech, the Sand Flea would have to be big enough to basically just roll over single-story buildings), but perhaps an explosive payload or something in the neighborhood of 10-20 lbs.

The Sand Flea would be able to monitor an area for far longer and is much quieter than a quadcopter. Also, the Sand Flea can enter buildings/caves, and likely carry a small munition load.

The REF doesn't invest in things that aren't ready to go to the sandbox. I think this is far further along than mid-stage.

All I could think of while watching it was 'its a parkour robot!'

Haha. Someone should do a web scrape of all parkour club directories, and then look at the intersection of that set with Boston Dynamics' team... could explain some things

Fantastic news, we are now one step closer to Pirate Bot / Wireless Internet Broadcasting robot that can circumvent internet shutdowns.

As regulations clamp down on internet free speech and authorities give themselves the power to administer internet blackouts during protests. This is the best way to respond in order to re-balance power back to democracies and people.

Now we need to make it a bit bigger and start working on the rechargeable battery pack, solar panel, kinect 3d camera, gps, and long range wireless router (the longest range wireless router allowable by law is a 1 watt router with a 1,200 foot range called Bountiful Router) but I'm sure you can extend the range much further with an illegal hacked version, and use repeater units to send signals further.

Basically a rogue mars-rover for Earth that sends out pirate signals. The future is horrifyingly awesome.

I wonder how it determines the height to see if it can leap or not.

I would think also it could be engineered to climb the side of the building by simply creating a triangle shape and motoring up the side or with some kind of suction.

I am assuming that its not autonomous and the angle/thrust is either pre-programmed for the demo or is human operator controlled.

I'm thinking just basic geometry and some rangefinders and other sensors. It needs to know how high off the ground its rangefinder is, and how far away the base of the target is. Once it knows that, it's some basic arithmetic to determine the height of its target landing area. Set a trajectory, charge for both the distance and height (and possibly to compensate for wind) and you're done.

There's some implementation subtleties I'm sure, especially considering your launch surface may vary, but that's the gist of it.

> Jumps of 1-8 m heights are user selectable

> Laser-based ranging to guide launch

(http://www.bostondynamics.com/img/SandFlea%20Datasheet%20v1_...)

Their RHex bot is pretty impressive too: http://www.youtube.com/watch?v=ISznqY3kESI

This is so awesome! I was a big fan of the Big Dog but this thing is slick and might be my new favorite robot!

We need to put a camera on this and send it to Mars

What, am I seriously the first one to say this? I don't care what this thing is supposed to be used for; I want one!

That's really cool and Boston Dynamics is amazing but there's most certainly not a 30ft jump in that video.

Now we just need a Clifford the Big Red Dog robot.

Doesn't the building jump look really fake to people? The image changes at 0:16 right before it pops up. I call bullshit.

I don't necessarily call bullshit, but I can't help but question what was cut out.

Reply on news.ycombinator.com