GITAI Partners With JAXA To Send Telepresence Robots To Space (ieee.org)

GITAI is a robotics startup with offices in Japan and the United States that's developing tech to put humanoid telepresence robots in space to take over for astronauts. IEEE: This week, GITAI is announcing a joint research agreement with JAXA (the Japanese Aerospace Exploration Agency) to see what it takes for robots to be useful in orbit, with the goal of substantially reducing the amount of money spent sending food and air up to those demanding humans on the International Space Station. It's also worth noting that GITAI has some new hires, including folks from the famous (and somewhat mysterious) Japanese bipedal robot company SCHAFT.

[...] GITAI says that their robots will "reduce the cost of space work to 10 percent" of the cost of using a real astronaut, by instead relying on earthbound humans for immersive teleoperation. As you might expect, the trouble with immersive teleoperation between Earth and orbit is getting data back and forth over a restrictive network. Part of GITAI's secret sauce involves compressing "data of 360-degree camera with resolution of 2.7K from original data volume of 800 Mbps to average 2.5 Mbps." At the same time, they've managed to reduce latency to 60 ms, which is really quite good, for talking to space. The plan is to get all of this working in low Earth orbit by 2020.

Comments

Missing the point

[Immerman]

>with the goal of substantially reducing the amount of money spent sending food and air up to those demanding humans on the International Space Station.

I'm embarrassed to see that level of missing-the-point from IEEE Spectrum, even as a joke (I hope). Robotic telepresence has nothing to do with reducing the human presence on the ISS - the ISS exists primarily to develop the technologies to keep humans alive and healthy in space (discovering the biggest problems being step one in that process).

Telepresence robots are about allowing us to expand our capabilities far beyond the ISS without the costs and risks of subjecting all the necessary workers to the risks. Especially by allowing people to work remotely in the hazardous environments in space, rather than having to go EVA. Perhaps some of the workers will remote in from Earth, though with a round-trip lag that increases to more than 133 milliseconds between surface and low orbit (the time taken for a light to travel halfway around the Earth and back again), they will substantially increase reaction times, and thus be unsuitable for anything that requires any kind of fast reflexes. But that still leaves a *lot* of jobs that could be done remotely, if slightly awkwardly. With practice you could probably compensate for the extra ~1/8th of a second of lag fairly well, especially if lag was artificially added so that it remained constant at the worst-case situation, rather than constantly changing.

They're hardly limited to orbital applications either - the moon would introduce a minimum of 2.6 seconds of lag for Earth-based operators, which would probably be too much to manually compensate for with anything that involves any reflexes - but you might be able to develop "AI reflexes" that can anticipate your intentions well enough for most situations. Or, you just have the operators safely inside a lunar habitat while they work remotely on the surface without worrying about radiation, oxygen, etc. Save the EVAs for recreation and especially difficult problems (though it seems to me that a good telepresence robot should be at least as dexterous as a person wearing bulky vacuum/radiation gloves.)

And of course, there's no reason a telepresence robot has to be bound by human limitations. A 10m tall robot torso with its "eyes" 60 cm apart would let an operator do large-scale assembly at an apparent 1/10th scale, without the awkwardness normally associated with operating banks of levers, etc. to control the dozens of degrees of freedom available.

Re:Missing the point

[rtb61]

Why protect those, who want the risk. Reality is the core stuff should be the focus, life support, better space drives, sturdier hulls. There are literally millions of people who very much want to take that risk.

Re:1.3 light seconds to the moon

[Immerman]

We can't. Ever. Not really. Not without rewriting the laws of physics. 1.3 light-seconds away means 2.6 seconds of unavoidable lag. You might be able to operate a "Diablo style" point-and-click RPG style interface serviceably with that kind of lag, micro-managing a limited-intelligence robot, but it will always be woefully inadequate for a "First Person" VR experience.

Not necessarily completely unserviceable, especially with 360* VR recording so that you can at least look around in "real time" (though 1.3 seconds into the past). Take things slow and easy, and as long as there's no chaotic interference on the moon, you could get things done. Eventually.

Much better though to have the operator *also* on the moon, comfortably secure within a nice safe habitat while they operate the same telepresence robot in real time.

Or, combine the two. A Moonie technician can be operating a "cyborg" body in real time, while a team on Earth offer guidance from a 2.6 seconds in the past.

Re:The space station was never about science

[Immerman]

Actually it is about science - but the real science is mostly studying the hazards of living in space, and how to mitigate them. The astronauts are highly-competent lab rats - the additional science they perform themselves is mostly just icing on the cake.

And we're not doing that "just because it's there", but because space offers an *enormous* font of wealth and opportunity, and the only possible place our species can survive in the long term. But first we have to develop the technology to survive there.