Want to help NASA drive the Mars Rover through AI?

Want to help NASA drive the Mars Rover through AI?

You can indeed do just that from home – no fine print involved

Do you really want to try your hand at astronomy? Why not start by helping NASA in driving Curiosity – their Mars Rover? That too on the Martian surface! All you need to do is to tag raw images with appropriate labels from the comfort of your home. And NASA would be grateful.

We all know about the rovers that NASA periodically sends to the moon, and now to a few nearby planets as well. They are wheeled mini-labs with different gadgets and appendages that are used to probe the surface and collect samples to understand the host better. But have you ever wondered how they move around on that distant and unknown terrain? And who drives them anyway? The answer is not really a secret. Scientists navigate them from earth, sitting in control rooms panelled from floor to ceiling with monitors, meters and consoles – as we so often see in sci-fi movies.

For the Mars Mission, they are currently training a terrain-classification algorithm called SPOC (Soil Property and Object Classification) to do a better job. It creates a visual map by analysing raw pictures sent from the Mars’ surface and labels the different terrain types visible on the images. Based on this map, mission scientists back at the control room identify the appropriate route to follow while navigating Curiosity on the Martian surface. Although the algorithm has been deployed, NASA is looking to refine it with more input data.

And there is no dearth of data, too. Because there are already 9000 images of Mars landscape to be classified and labelled, while more keep pouring in. Someone has to label and annotate every landscape feature on every image so that the algorithm can learn them better. And this is where NASA is asking for crowdsourced help.

All raw images have been uploaded on Zooniverse – a citizen science web – from where anyone from anywhere in the world can access them and do the task of labelling after going through the online tutorial provided by NASA. Currently, the work can be done only in English – but Spanish, Hindi, Japanese and several other language-versions are on its way, so that more people from all over the globe can do their bit.

There is no compulsion, or obligation. It is a voluntary initiative for the cause of expanding human scientific knowledge and even a few minutes of work each day would be of immense help.

If you are interested, please visit the original announcement from NASA at the following link:


Leave a comment

Your email address will not be published. Required fields are marked *

© 2024 Praxis. All rights reserved. | Privacy Policy
   Contact Us