UK Markets closed
  • FTSE 100

    +59.28 (+0.85%)
  • FTSE 250

    +206.11 (+0.91%)
  • AIM

    +4.45 (+0.36%)

    -0.0003 (-0.03%)

    -0.0007 (-0.0495%)

    -343.05 (-1.45%)
  • CMC Crypto 200

    -17.83 (-2.25%)
  • S&P 500

    +43.04 (+0.99%)
  • DOW

    +236.50 (+0.68%)

    +0.02 (+0.03%)

    -2.90 (-0.16%)
  • NIKKEI 225

    +159.80 (+0.58%)

    -401.86 (-1.45%)
  • DAX

    +154.75 (+1.00%)
  • CAC 40

    +87.23 (+1.35%)

Scientists train drone to hunt for meteorites that have crashed to Earth

·2-min read
 (Getty Images)
(Getty Images)

Researchers have trained a drone to search for asteroids that crash to Earth.

The drones autonomously fly in a grid pattern, taking photos of the ground over a large area, and use artificial intelligence to search through the pictures to identify potential meteorites.

Every year approximately 500 meteorites fall to the planet’s surface, but less than two per cent of these are ever recovered – sometimes because they fall in inaccessible locations, such as the ocean, but in other instances simply because they are not found in time.

The drone, developed by the University of California, Davis, has been tested around Walker Lake in Vegada, where a meteorite fell in 2019. “Images can be analyzed using a machine-learning classifier to identify meteorites in the field among many other features," said Robert Citron, a postdoctoral researcher at the university.

The proof-of-concept meteorite classifier deploys a combination of "different convolution neural networks to recognise meteorites from images taken by drones in the field," his team says in a Meteoritics & Planetary Science paper.

The software has not yet been perfectly refined. While it was able to identify test meteorites placed on a dry lake bed, it also brought up a number of false positives – but the researchers say they are optimistic about the progress made.

"Since a future meteorite fall could occur on any terrain," Citron told said Universe Today, "the system needed an object detection algorithm trained with examples of many types of meteorites on various terrain types. To create a properly trained object detection network, thousands of example images are required.

"Fortunately, every field test we gain more data that we can incorporate into the dataset and use to retrain the object detection network and improve the accuracy," Citron said. "So, we will continue to try and improve the detection accuracy. Currently we need a better drone with a higher resolution camera."

It takes approximately 100 hours to find one meteorite through conventional methods, so the ability to shorten that time through technology could save researchers vast amounts of work.

If a meteorite can be recovered, the light curve and deceleration profile – gathered through analysis of the rock – can be used to improve predictions about the altitude of where the asteroid came from.

With over 40 asteroid families in the asteroid belt, such discoveries could help scientists unlock the secrets of the early universe and gather more information about the evolution of our solar system.

Read More

Euro 2020: What could social media companies actually do about racist abuse – and would it work?

Elon Musk shares poem about people who ‘attack space’ amid race between billionaires

Bitcoin price – live: Crypto market is ‘ticking time bomb’ amid Apple rumours

Our goal is to create a safe and engaging place for users to connect over interests and passions. In order to improve our community experience, we are temporarily suspending article commenting