NASA Asteroid Challenge: Develop an algorithm to detect asteroids

[Unity+]

I think this is the best place to post this. I hope it isn't too late. It might have been posted before.

 

 

 

Managed by the NASA Tournament Lab, the entire contest series runs through August and is the first contest series contributing to the agency’s Asteroid Grand Challenge.

“For the past three years, NASA has been learning and advancing the ability to leverage distributed algorithm and coding skills through the NASA Tournament Lab to solve tough problems," said Jason Crusan, NASA Tournament Lab director. "We are now applying our experience with algorithm contests to helping protect the planet from asteroid threats through image analysis.”

The Asteroid Data Hunter contest series challenges participants to develop significantly improved algorithms to identify asteroids in images captured by ground-based telescopes. The winning solution must increase the detection sensitivity, minimize the number of false positives, ignore imperfections in the data, and run effectively on all computer systems.

“Protecting the planet from the threat of asteroid impact means first knowing where they are,” said Jenn Gustetic, Prizes and Challenges Program executive. “By opening up the search for asteroids, we are harnessing the potential of innovators and makers and citizen scientists everywhere to help solve this global challenge.”

Gustetic and Jason Kessler, Grand Challenges Program executive, will host a panel March 10 at the South by Southwest Festival in Austin, Texas titled “Are We Smarter than the Dinosaurs?” to talk about how open innovation can meaningfully engage people in discussions on and research into space exploration and help us solve problems of global importance. They will provide an outline of the Asteroid Data Hunter contest series and other efforts to detect asteroid threats, as well as ideas for mitigating these threats.

“Current asteroid detection initiatives are only tracking one percent of the estimated objects that orbit the Sun. We are excited to partner with NASA in this contest to help increase the quantity and knowledge about asteroids that are potential threats, human destinations, or resource rich.” said Chris Lewicki, President and Chief Engineer of the asteroid mining company Planetary Resources, Inc. “Applying distributed algorithm and coding skills to the extensive NASA-funded Catalina Sky Survey data set will yield important insights into the state of the art in detecting asteroids.”

Through NASA's asteroid initiative, 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 Grand Challenge, one part of the asteroid initiative, expands the agency's efforts beyond traditional boundaries and encourages partnerships and collaboration with a variety of organizations.

The algorithm contests are managed and executed by NASA's Center of Excellence for Collaborative Innovation (CoECI). CoECI was established at the request of the White House Office of Science and Technology Policy to advance NASA open innovation efforts and extend that expertise to other federal agencies. CoECI uses the NASA Tournament Lab (NTL) for its advanced algorithmic and software development contests. Through its contract with Harvard Business School in association with Harvard's Institute of Quantitative Social Science, NTL uses the topcoder platform to enable a community of more than 600,000 designers, developers and data scientists to create the most innovative, efficient and optimized solutions for specific, real-world challenges faced by NASA.

http://www.nasa.gov/content/be-an-asteroid-hunter-in-nasas-first-asteroid-grand-challenge-contest-series/#.UyAY0fmwLGE

 

I, myself, have programming experience in many languages. If there are some astronomers out there that would be willing to work with me and other programmers that are out there, we could use our knowledge to develop an algorithm that could detect asteroids simply using ground-based telescopes. We can also use physics knowledge to solve this problem. I don't know if this contest is already over, but if it is and no one has solved it yet then I think we could take up the challenge.

Edited March 12, 2014 by Unity+

[Ophiolite]

I don't know if this contest is already over, but if it is and no one has solved it yet then I think we could take up the challenge.

The potential threat of NEO impacts is a very serious one. I am pleased that governments, organisations and individuals around the world take the issue seriously. I applaud your interest in this competition and wish you luck finding collaborators.

 

I am puzzled by your uncertainty about whether or not the contest is over. The NASA release is dated March 10th and the text states,

 

"NASA’s Asteroid Data Hunter contest series will offer $35,000 in awards over the next six months to citizen scientists who develop improved algorithms that can be used to identify asteroids."

 

and

 

"The first contest in the series will kick off on March 17."

 

and

 

"Managed by the NASA Tournament Lab, the entire contest series runs through August and is the first contest series contributing to the agency’s Asteroid Grand Challenge."

 

[Unity+]

The potential threat of NEO impacts is a very serious one. I am pleased that governments, organisations and individuals around the world take the issue seriously. I applaud your interest in this competition and wish you luck finding collaborators.

 

I am puzzled by your uncertainty about whether or not the contest is over. The NASA release is dated March 10th and the text states,

 

"NASA’s Asteroid Data Hunter contest series will offer $35,000 in awards over the next six months to citizen scientists who develop improved algorithms that can be used to identify asteroids."

 

and

 

"The first contest in the series will kick off on March 17."

 

and

 

"Managed by the NASA Tournament Lab, the entire contest series runs through August and is the first contest series contributing to the agency’s Asteroid Grand Challenge."

 

Well, thank you for pointing those out. I apologize for my stupidity on that.

[Ophiolite]

Not a problem.

 

I'm not clear on what you are planning when you suggest "using an algorithm". Could you elaborate? I think there are well established methods for computing orbits of NEOs once these are identified. The problem lies, I thought, in detecting them in the first place and then in observing them over a long enough period, or a sufficient number of passes, to pin down the orbital parameters. Have I got that wrong? (This is not my area of expertise. I'm more interested in what happens when they hit.)

[Unity+]

Not a problem.

 

I'm not clear on what you are planning when you suggest "using an algorithm". Could you elaborate? I think there are well established methods for computing orbits of NEOs once these are identified. The problem lies, I thought, in detecting them in the first place and then in observing them over a long enough period, or a sufficient number of passes, to pin down the orbital parameters. Have I got that wrong? (This is not my area of expertise. I'm more interested in what happens when they hit.)

Being able to predict orbits of asteroids is a part of detecting them. A part of the problem is finding what are asteroids and what are other celestial objects. Being able to more accurately predict the movements of asteroids would help detect asteroids with less and less errors and false alarms.

 

EDIT: Adding onto this, being able to find what movements asteroids will make will help to use ground-based telescope imagery to detect asteroids more accurately.

Edited March 12, 2014 by Unity+

[Unity+]

If you want to collaborate with me on a team for this kind of project, please PM me so we can set up a group for this project.