Artificial intelligence (AI) is transforming numerous industries, and space exploration is no exception. As we venture deeper into the cosmos, AI becomes more critical in overcoming the challenges of long communication lags, handling massive datasets, and enabling autonomous robotic planetary exploration systems.
Image Credit: Frame Stock Footage/Shutterstock.com
Processing Extreme Data Volumes
The exponential rise in space data captured from satellites, telescopes, and interplanetary probes requires AI's analytical capabilities. Modern space instruments generate terabytes of data daily - far more than scientists can examine manually.
AI automation assists with classifying and processing streams of images, sensor readings, and spectral data. For example, NASA employs AI in the Mars Reconnaissance Orbiter, which uses AI techniques to filter and prioritize over six megabits per second of data. Scientists trained these AI algorithms to recognize key features from billions of images of Mars' surface.
Furthermore, astronomers employ AI to scour astronomical datasets. Scientists trained neural networks to identify exoplanets from dips in light curves captured by the Kepler space telescope. These AI tools also recognize and classify galaxy types and cluster stars based on shared motion.
NASA collaborated with Google to train extensive AI algorithms to analyze data from the Kepler exoplanet mission, leading to the discovery of two new exoplanets, Kepler-90i and Kepler-80g, that scientists had previously overlooked. This success prompted the use of AI on NASA's TESS mission data to identify candidate exoplanets.
"New ways of looking at the data – such as this early-stage research to apply machine learning algorithms – promise to continue to yield significant advances in our understanding of planetary systems around other stars. I'm sure there are more firsts in the data waiting for people to find them." Jessie Dotson, NASA Ames Research Center's Kepler project scientist.
In research published in Astronomy and Astrophysics, led by University of Leeds' researcher Miguel Vioque, AI was introduced in the Gaia space telescope data analysis. This led to the discovery of 2,000 protostars, a substantial improvement from scientists' previous identification of only about 100 stars before adopting AI and machine learning techniques.
AI holds immense promise for automating spectral data analysis from future missions to places like Saturn's moon Enceladus, where rapid onboard processing will be essential to identify possible signs of microbial extraterrestrial life in ice plumes emanating from a subsurface ocean.
Enabling Autonomous Robotic Planetary Exploration
AI empowers robotic rovers on planetary surfaces like Mars by providing advanced autonomy for tasks such as vision-based navigation, path planning, object detection, and adaptive mission prioritization, enabling them to traverse rugged and unfamiliar terrain using onboard maps and sensor data.
For example, NASA's Curiosity and Perseverance rovers leverage AEGIS, a powerful AI system, to build autonomous 3D terrain maps and identify rock features and soil composition. It can even recommend the day's activities based on terrain complexity, energy usage and scientific value.
Such intelligent capabilities will become vital as future rover missions target more distant destinations with greater communication delays from Earth, like gas planets and their icy moons. In addition, AI enables autonomous navigation and adaptable science; rovers can respond to discoveries immediately rather than awaiting delayed commands.
AI also assists with entry, descent, and landing (EDL) - the riskiest phase facing probes sent to Mars. The autonomous guided entry capabilities pioneered by the Mars Science Laboratory enable trajectory correction by comparing real-time sensor data against high-resolution surface maps to reach designated landing zones accurately.
As agencies plan more ambitious robotic missions, AI provides the advanced autonomy to explore harsh and alien environments.
Supporting Astronaut Health
The mental and physical toll during multi-year missions creates a need for enhanced astronaut medical care. AI shows promise for improving future crew support systems.
By integrating multi-modal data streams - from sensors tracking heart rate and skin temperature to recording exercise and sleep patterns - predictive health analytics powered by AI can enable customized interventions tailored to each astronaut. Combining real-time vital signs, behavioral indicators, and environmental conditions holistically allows sophisticated diagnostics, early risk warnings, and personalized treatment plans.
For example, the Crew Interactive Mobile Companion (CIMON), designed by Airbus, IBM, and the German Aerospace Center, is a voice-controlled AI robot that flew to the International Space Station (ISS) in 2018.
CIMON can see, hear, understand, and speak via voice and facial recognition, enabling it to navigate the space station, locate and retrieve items, document experiments, and display procedures.
Most importantly, CIMON acts as an emotive consoling companion that can sense stress levels. It has been trained in psychological support using Watson's natural language capabilities and can lead astronauts through therapeutic experiments to elevate mood.
The ISS and lunar Gateway will test further intelligent systems to anticipate astronauts' needs, provide recommendations, and automate routine activities. AI virtual assistants are also being adapted for psychological support on future Mars missions facing communication lags when contacting ground control.
Conclusion
AI's role in space exploration is transformative, processing vast data from celestial bodies and predicting risks like solar storms and space debris. It enhances spacecraft autonomy, reduces human reliance, and aids astronauts with operations, navigation, and satellite monitoring.
More from AZoQuantum: Modified Newtonian Dynamics and the Search for a Ninth Planet
References and Further Reading
AM, L. (2023). Space missions out of this world with AI. https://doi.org/10.1038/s42256-023-00643-3
Chien, S., & Morris, R. (2014). Space applications of artificial intelligence. Ai Magazine, 35(4), 3-6. https://doi.org/10.1609/aimag.v35i4.2551
ESA. (2023). Artificial intelligence in space. [Online]. Available at: https://www.esa.int/Enabling_Support/Preparing_for_the_Future/Discovery_and_Preparation/Artificial_intelligence_in_space
Garanhel, M. (2022). AI in space exploration. [Online]. Available at: https://www.aiacceleratorinstitute.com/ai-in-space-exploration/
Kumar, S., & Tomar, R. (2018, February). The role of artificial intelligence in space exploration. In 2018 International Conference on Communication, Computing and Internet of Things (IC3IoT) (pp. 499-503). IEEE. https://doi.org/10.1109/IC3IoT.2018.8668161
Loeffler, J. (2023). Deep space missions will test astronauts' mental health. Could AI companions help? [Online]. Available at: https://www.space.com/astronauts-artificial-intelligence-companions-deep-space-missions
Marr, B. (2023). Artificial Intelligence in Space: The Amazing Ways Machine Learning Is Helping to Unravel the Mysteries of The Universe. [Online]. Available at: https://www.forbes.com/sites/bernardmarr/2023/04/10/artificial-intelligence-in-space-the-amazing-ways-machine-learning-is-helping-to-unravel-the-mysteries-of-the-universe/?sh=303f015f7b60
NASA. (2017). Artificial Intelligence, NASA Data Used to Discover Eighth Planet Circling Distant Star. [Online]. Available at: https://www.nasa.gov/news-release/artificial-intelligence-nasa-data-used-to-discover-eighth-planet-circling-distant-star/
NASA. (2023). Mars Reconnaissance Orbiter - Telecommunications. [Online]. Available at: https://mars.nasa.gov/mro/mission/spacecraft/parts/telecommunications/
Russo, A., & Lax, G. (2022). Using artificial intelligence for space challenges: A survey. Applied Sciences, 12(10), 5106. https://doi.org/10.3390/app12105106
Vioque, M., Oudmaijer, R. D., Schreiner, M., Mendigutía, I., Baines, D., Mowlavi, N., & Pérez-Martínez, R. (2020). Catalogue of new Herbig Ae/Be and classical Be stars. A machine learning approach to Gaia DR2. arXiv preprint arXiv:2005.01727. https://doi.org/10.1051/0004-6361/202037731
Disclaimer: The views expressed here are those of the author expressed in their private capacity and do not necessarily represent the views of AZoM.com Limited T/A AZoNetwork the owner and operator of this website. This disclaimer forms part of the Terms and conditions of use of this website.