Hacker News Reader

Comment by aidanhs | Hacker News Reader

Also an impressive achievement on the software side of things - no room for showstopper bugs here! It's not quite as simple as taking it offline, deploying some new code and starting it up again.

Is there any information on the processes used to have such a high level of confidence in the software? Lots of simulation? Relatedly, it might also be interesting to see a list of 'known bugs' and ways they've been worked around.

View on HN

http://cacm.acm.org/magazines/2014/2/171689-mars-code/abstra... is a good watch.

And you can get a fair bit of insight through many of the videos from Flight Software Workshop - http://flightsoftware.jhuapl.edu/files/_site/workshops/2014/

AFAIK it's similar to the processes for avionics software, but with even more attention to detail. This link came up recently on HN in reference to avionics:

https://en.wikipedia.org/wiki/DO-178B

Hmm, I am not sure if they would have very sophisticated software on the lander, in the way your thinking. I don't think any processing is done on the lander itself. It might be just primarily used as a remote sensing telemetry system. I suppose the lander would have a lot of power management circuits, some control circuits, sensor-actuator circuits, and transmitter-receiver communication circuitry. There would probably be a RTOS for real-time measurement and actuator motion.

Still, I think someone with specific expertise might give a better answer.

> Hmm, I am not sure if they would have very sophisticated software on the lander, in the way your thinking.

I think it's safe to say that's still an order of magnitude more sophisticated than anything I deal with on a daily basis ;)

Reply on news.ycombinator.com