NASA launches $35,000 bounty for improved asteroid detection algorithms

I'm happy to see NASA continuing its tradition of reaching beyond traditional communities to work on hard problems (e.g., its prior Glove challenge). Used correctly, these competitions hold the promise to help engage the public, source novel innovations, and save the government money (plus be fun!).

BUT, as somebody who recently participated in a DoD-sponsored contest that was rife with improprieties, bias, and incompetence (the government effectively threw away the rule book in making the reward), please be careful in reviewing the rules and verifying that the administrators of this are accountable. Regardless of if the amount at stake is $50K or $1M, bureaucrats have an astonishing capacity for rationalizing whatever outcome they choose or stumble into.

A related video, showing orbits of asteroids discovered over time: https://www.youtube.com/watch?v=xJsUDcSc6hE

It starts with about 10k asteroids. By the end there's 10k being added every month.

"... This approach will discover new asteroids and improve our model of the solar system. It also opens opportunities for interesting algorithmic challenges, and the chance for a normal person to discover an asteroid.

"As a programmer, I’m particularly interested in how we’ll improve algorithms to spot asteroids. I am sure that there are conventional ML techniques and even simpler image processing approaches that are not being sufficiently exploited." [1]

That's Ian Webster talking about Asterank [2], which might be relevant background here since this NASA project is in partnership with Planetary Resources, which acquired Asterank.

1. http://www.ianww.com/blog/2013/11/26/asterank-discover-revie...

2. http://www.asterank.com/discover

Maybe I'm just a bit too cynical (negative?), but I find it rather sad that, lets say to improve a movie recommendation algorithm we have a $1M bounty (Netflix price), while to improve an algorithm to detect the possible destruction of earth we have a bounty of $35K.

Just use dinosaurs...

Algorithm or technique? The press release is quite vague (I did not check through the links). Are we talking laser, infrared, visible spectrum, or fruity pebbles algos? What if a pure black object is occluding known stars? What if there is some ice on the same black object? Is the algo meant to be run directly on a satellite (Hubble, Kepler, etc.) or from a terrestrial observatory?

Break it into knowns, then unknowns.

k0

> the agency seeks to enhance its ongoing work in the identification and characterization of near-Earth objects for further scientific investigation. This work includes locating potentially hazardous asteroids and identifying those viable for redirection to a stable lunar orbit for future exploration by astronauts.

The asteroid redirection idea sounds really awesome and incredibly complicated. Is it what NASA plans to do in case we have one heading straight at us? Do we have to blow up those that can't be redirected? What exactly does “viable for redirection” mean?

well here's a way of detecting asteroids: http://www.atari.com/arcade/asteroids can I get the $35k? :-p