From understanding our origins to predicting the upcoming events, we have made some of the greatest breakthroughs by studying the universe on earth. By means of artificial intelligence and machine learning, tech companies are accelerating this kind of research at an unprecedented pace.
Pony Ma Huateng, the founder and CEO of Tencent Holdings, announced the company’s collaboration with the National Astronomical Observatory (NAOC) to search for new stars at the 2021 World Artificial Intelligence Conference in Shanghai on Thursday.
The "Star Exploration Project" jointly launched will be based on the computer vision algorithm of Tencent Yutu lab and the computing capacity of Tencent Cloud. Leveraging AI, the project can process abundant data received from China's Five-hundred-meter Aperture Spherical radio Telescope (FAST) and find pulsar clues through visual AI analysis.
The data generated by FAST in one week are approximately 30 million images which would take a whole year by handpicking without any break, but it only takes 3 days to complete after AI pre-screening. The combination of AI technology will greatly save labor and time costs.
Our ancestors used the Big Dipper to navigate in the dark night for its distinctive pattern and Pulsars have similar characteristics. Since the gamma-rays of a spinning pulsar are beamed like a lighthouse beacon at a unique frequency, the pulsar research is helpful for positioning in the universe and is expected to bring answers to many major astrophysics’ questions.
In exploring extraterrestrial space, the world is making continuous efforts. In 2020, Google Cloud said that the Vera C. Rubin Observatory will use its cloud services to host its Interim Data Facility, which will capture 1,000 images of the sky every night for 10 years and collect astronomical data.In the field of finding other planets in the universe, NASA employed the Planetary Spectrum Generator (PSG) to create 3-D orbits and atmospheric properties of the exoplanets for experimentation.
"In the past year, AI has brought more and more convenience to life in areas such as medical care, urban governance and non-contact services. But in the field of AI, what is unknown remains greater than the known. We must make AI technology knowable, controllable, usable and reliable. We have to outsmart the increasingly powerful AI technology to keep it in hand," said Ma.
At present, there are professional manufacturing enterprises in Shanghai that use Tencent's AI technology for quality inspection, in a bid to achieve cost reduction and efficiency increase, Ma added.