Hacker News Reader

Comment by bArray | Hacker News Reader

If only they had the robotics we have today for the initial clean-up process. I'm not saying the robots wouldn't have died from radiation, but better a robot discover a high source of radiation to its peril than a human.

It seems clear to me that autonomous systems will be the only way to go in such scenarios - most types of external communication are near impossible and the situation itself can be dynamic.

I've occasionally seen some very awesome work come out of Japan and the Fukushima nuclear cleanup process. Fingers crossed we can advance robotics far enough for the devastating scenario where they are really needed.

View on HN

They did use robots leftover from the Lunokhod lunar exploration program and a robot sent by West Germany, however the radiation level was so high that they failed almost immediately. They had to resort to using "biorobots", as they cynically called the soldiers used to clean up the site. I'm not sure if modern robots would fare any better due to their higher complexity and use of lots of electronics which is sensitive to radiation.

I don't know if it possible to have advanced robots is such high-rad environments. The inspector-bots in Fukushima, although supposedly rad-hardened, all failed quickly. I agree it would be ideal if possible.

https://www.theverge.com/2017/2/17/14652274/fukushima-nuclea...

You mentioned good work coming from the Fukushima site cleanup effort. Do you have any links? I'd love to read about what successes they've had after the initial problems with the ones discussed above.

I think that with modern fiber-optics it might be possible to make a really long robot arm without any sensitive circuits at the extremity.

I'm not sure how fibre optics could help here, an you expand?

They're describing running the computers 100's of meters away with fibre to the robot. But it won't help with motor encoders and sensors that all must be onboard.

Interesting idea though. To some degree is probably possible. No such thing as a rad hardened cmos though

Why couldn't the necessary electronics be shielded with a shroud of lead? It's not like these robots have lunar weight requirements.

Spalling/scatter. In certain radiation environments, shielding makes things worse, as an energetic particle can cause a knock-on flood of even more damaging particles to be emitted if strikes an atom in the shield. Ultimately more mass does help, but we're talking meters-thick, not just another few cm.

The details are a lot more complicated than that. It's just worth keeping in mind that for any given radiation environment, there may be times when either more or less shielding would be a better choice, as counterintuitive as that may seem.

I think the idea is to have a passive bundle of fiber at the far end with all your electronics in the safe zone, like a fiber optic endoscope.

I’m curious. Which of the parts back in the day would have failed first and why? And which of the parts would have continued functioning despite radiation?

Computers fail first because the memory content gets corrupted. I can imagine, that power electronics will continue functioning despite radiation.

Apparently regular MOSFETs are affected[1] by the total dose of ionizing radition (TDI) causing important parameters to change, such as threshold voltage (Vgt) and internal resistance (Rds).

So it seems even power electronics are not safe[2], nor microcontrollers/microprocessors and other types of ICs.

[1]: https://nepp.nasa.gov/files/28954/2017-Lauenstein-NSREC-Post... (esp. figure 11)

[2]: https://www.renesas.com/eu/en/products/space-harsh-environme...

Thank you, signed for Renesas newsletter. Very interesting topics.

It isn't quite possible to create a robot capable of working in 15,000-20,000 r/h fields. Less so in 1986.

Reply on news.ycombinator.com