Today a black hole was observed … avoiding a pothole tomorrow?
On Wednesday, researchers from Project Event Horizon Telescope released the first images ever taken of a black hole – a space-based gravitational hole that is strong enough to light itself up. Taking these pictures was an amazing achievement. But the technologies developed to produce images of supermassive emptiness 55 million light-years away could have far-reaching effects on the Earth. This also applies to the people who developed it.
First of all, Albert Einstein theorized that black holes were only detected through the gaps in our data. Scientists announced that in 2001 the Hubble Space Telescope observed how ultraviolet light faded and finally disappeared when it fell into the Cygnus XR-1
To get a picture of the black hole itself, the EHT project used a network of ten earthbound radio telescopes that were interconnected and acted as a single system. The telescopes collected high-frequency radio waves from space, and four independent teams of scientists used algorithms to transform the radio signals into visual images.
Coordinating telescopes and collecting radio signals was complex and impressive, but the development of the algorithm is likely to have a long-term effect on the technology, said Jonathan Weintroub, an electrical engineer who developed physical tools for the EHT project. His team used standard products and existing telescopes to essentially build a world-spanning telescope, just as a child could build a Lego model. That is no small matter. The final system was able to collect and store 5 petabytes of data. If 1 byte were a 2 foot by 2 foot field, 1 petabyte would cover the entire earth. However, to turn that data into an image, completely new software tools had to be created.
The problem: This global mega-telescope (though obviously great) still provides such incomplete data as a piece of Swiss cheese. The telescopes collect photons – packets of light – that fall like proverbial pennies from the sky out of space. But even if they work together, they can only catch a tiny sample of these photons. Reconstructing an image from this sparse dataset poses a challenge as massive as the black hole itself, Weintroub said. The algorithms developed by the EHT researchers were crucial to solving this challenge, and their solution could have far-reaching implications.
Imagine trying to put together a puzzle that misses 90 percent of the pieces. Not only is it difficult to get the picture right, it's also hard to know which image you want to create. "Because we have such sparse measurements, there are an infinite number of images that could match the data," said Lindy Blackburn, a MIT student who works as a data scientist on the EHT project.
EHT scientists' algorithms helped to contain the infinite number of possible images by finding out which results were physically plausible and unlikely. Blackburn, for example, told me that the algorithms all tended to favor the images that could easily explain the measurements of the telescopes and sort out the images with many fine details or complicated features. When applying such a list of constraints, putting the puzzle together was (or is at least realistic) a little less harsh. It's not perfect – the image of the black hole is blurry, Blackburn told me, in part because each of the four teams produced a slightly different image and the researchers were reasonably conservative about what details made them the final, representative image. But it was enough to turn radio-wave data into a picture.
And that's what matters, Blackburn said, because astronomy is not the only field that faces the problem of converting sparse data into images. For example, in medical imaging, doctors use MRI to transform radio waves into images of your body. It is also an integral part of self-driving cars that rely on computer visualization to "see" everything from pothole to man. The types of algorithms that have been developed to photograph a black hole built on this research from other fields and could in turn help to improve the view of computers on Earth. This blurry picture of a dark vortex in space could end up as a chapter in the story of how man could be safely driven by technological developments into 2-ton metal and plastic hunks powered only by computers.
If this is the case, Weintraub said It's not just the technology that changes the future, but also the people who made it. Many of the people working on imaging technologies for the EHT project are like Blackburn doctoral students. Photographing a black hole did not just mean developing cool technology – it meant enabling a number of young scientists to develop different ideas and being really good at developing new tools before they dispersed into science and industry. Blackburn staff member Katie Bouman, for example, worked as a postdoctoral fellow at the MIT Computer Vision Laboratory and worked on improving algorithms in many different areas and in computer vision applications. As part of the team that developed the eyes of the EHT, she will begin her first professorship at Caltech this fall. If we consider what a black hole may have seen for humanity for the first time, the role of the project as an incubator for scientific talent could be the biggest contribution.