Five ways artificial intelligence can help space exploration

The progress in artificial intelligence capabilities has allowed us to advance in all kinds of disciplines and not just those intended for applications on Earth. Here are some ways in which artificial intelligence can help us in space exploration - from mission planning to cleaning the orbit of space debris around the Earth

By Deep Bandyodkar, PhD Student at Aerospace Center of Excellence, University of Strathclyde, and Audrey Barkwand PhD Student in Mechanical and Aerospace Engineering, University of Strathclyde

Artificial intelligence has been making waves in recent years, allowing us to solve problems faster than traditional computing could ever allow. Recently, for example, Google's subsidiary specializing in artificial intelligence, DeepMind, developed AlphaFold2, a software that solved the protein folding problem that has occupied scientists for 50 years.

The progress in artificial intelligence capabilities has allowed us to advance in all kinds of disciplines and not just those intended for applications on Earth. Here are some ways in which artificial intelligence can help us in space exploration - from planning missions to cleaning up the orbit of space debris around the Earth.

Astronaut assistants

 CIMON will assist astronauts on the International Space Station. NASA / Kim Shifflett, CC BY

Robots like Terez and Case, the helper robots from the movie "Interstellar", do not yet exist in real space missions. But researchers are in the stages of developing something similar - creating intelligent assistants to help astronauts. AI-based assistants could be very useful for space exploration, although they won't look as fancy as the ones in the movies.

A recently developed virtual assistant can detect any dangers in long space missions such as changes in the composition of the air in the spacecraft - for example, excess carbon dioxide or a sensor malfunction that could be harmful. The virtual assistant will alert staff and provide suggestions for testing.

An AI assistant named Cimon was flown to the International Space Station (ISS) in December 2019, where it will be tested for three years. Ultimately, marking will help reduce the stress on the astronauts by performing routine tasks.

Space mission planning

Planning a mission to Mars is no easy task, but artificial intelligence can make it a little easier. New space missions traditionally rely on knowledge gathered from previous flights. However, this information can often be limited or not fully accessible.

This means that the flow of technical information is limited by the availability of those who can access it and share it with the other design engineers. But what if all the information from all previous space missions would be available to anyone with appropriate authorization in a few clicks? One day there may be a smarter system - similar to Wikipedia but with artificial intelligence that can answer complex queries with reliable and relevant information - to help in the early planning of new space missions.

The researchers are working on the idea of ​​an engineering assistant to shorten the time required to plan the initial task which currently requires many man-hours. 'Daphne' is another example of an intelligent assistant for planning earth observation satellite systems. Daphne serves as a systems engineer in satellite planning groups. It makes their work easier by providing access to relevant information, including feedback and answers to specific queries.

Satellite data processing

Observation satellites generate enormous amounts of information. This information is received by ground stations in parts over a long period of time, and must be joined together so that they can be analyzed. In the past, researchers used crowdsourcing projects to analyze basic satellite imagery at a very small scale. Artificial intelligence can help analyze detailed satellite data.

Artificial intelligence is very effective in processing such data intelligently. It could be used to assess hot spots in urban areas and combine meteorological data with satellite images to assess wind speed. AI has also helped estimate solar radiation using geostationary satellite data, and many other applications.

AI for data processing can also be used in the satellites themselves. In a recent study, scientists tested different AI techniques for a remote satellite health monitoring system. These systems are able to analyze data received from satellites to identify problems, predict satellite health performance and present a model for informed decision making.

space debris

One of the biggest space challenges of the 21st century is how to deal with space debris. According to the European Space Agency, ESA, there are almost 34,000 objects larger than 10 cm that pose a serious threat to the existing space infrastructure. There are some innovative approaches to dealing with the threat, such as designing the holiday satellites in low orbit so that they reenter the Earth's atmosphere and burn up in a controlled manner.

Another approach is to prevent possible collisions in space, to prevent the creation of space debris in advance. In a recent study, researchers developed a method for planning collision avoidance maneuvers using machine learning (ML) techniques.

Another innovative approach is to use the enormous computing power available on Earth to train ML models, transmit these models to a spacecraft already in orbit or on its way to it and use them also on the satellite itself to make various decisions. One way to ensure the safety of space flights has recently been proposed using networks already trained on spacecraft. This allows greater flexibility in the design of the satellite while reducing the danger of an orbital collision.

navigation systems

On Earth, we are used to tools such as Google Maps that use GPS or other navigation systems. But there is no such system for other bodies in space for now.

We don't have navigation satellites around the Moon or Mars, but we can use the millions of images we have from observation satellites such as the Lunar Reconnaissance Orbiter (LRO). In 2018, a team of researchers from NASA in collaboration with Intel developed a smart navigation system using artificial intelligence to explore the planets. They trained the system on millions of photographs available from various missions and created a virtual lunar map.

As we continue to explore the universe, we will continue to design ambitious missions to satisfy our curiosity as well as improve human life on Earth. In our endeavor, artificial intelligence will help us both on Earth and in space to make this research possible.

to the article on The Conversation website

More of the topic in Hayadan:

• A robot will assemble a modular space telescope in orbit
• A new robotic system will repair satellites in space
• Elon Musk's space settlement vision