Now that we are so accustomed to the technology, it may seem hard to live without GPS – for years, it has made it simple for us to locate new locations and to commute. Without the GPS’s reliable directions, can many of us even locate the closest store or theatre?

Working to create a GPS system that can be used to navigate the Moon, a NASA scientist is betting that the same advantages would apply in space.

An artificial intelligence (AI) system is being trained by Dr. Alvin G. Yew, a research engineer at NASA’s Goddard Space Flight Centre in Greenbelt, Maryland, to recognise landmarks on the Moon’s horizon. This AI system will be utilised by astronaut vehicles’ on-board navigational systems to determine their location on the lunar surface.

The ability to use a local map would be just as helpful for a robotic trip as it would be for a human one, according to Yew in a press statement. This is because the Moon’s thin atmosphere affects the visibility of landmarks, making it difficult to determine how far or how large distant mountains or other features may be.

Yew has a background in aerospace engineering and has worked on hardware development for missions including Astro-H, MMS, GPM, and FASTSat at the Goddard Space Flight Centre. He has also led numerous research projects to create and test ground-breaking space travel technology.

According to his bio, he specialises in “Guidance, Navigation, and Control components on small satellite architectures.”It has been involved in creating cutting-edge equipment for planetary science research, including as zero-gravity parabolic flights and early-stage ice sample techniques.

Backing up LunaNet

As NASA develops the navigational and communication capabilities for potential lunar colonies, Yew’s research also examines how it looks to create systems that can act as redundancies for one another.

By creating a network of lunar satellites for use by spacecraft and astronauts in establishing a long-term presence at or on the Moon, the Luna Net programme aims to provide internet and location services there.

Luna Net, a collaboration between NASA and other space organisations like the ESA and JAXA, would enable communication and data sharing throughout the entire Moon up to an altitude of 125 miles (200 km).

According to Cheryl Gramling, a NASA feature article’s Associate Chief for Technology of the Mission Engineering and Systems Analysis Division:

By guaranteeing that crewed and robotic missions can rapidly and precisely determine their locations and pass that information forward to their planning systems, Luna Net will “provide a new paradigm for Earth-independent navigation.”

Other features of LunaNet would include the ability to use space weather instruments to identify potentially dangerous space activity, such as solar storms, and send users immediate warnings without waiting for guidance from network administrators on Earth.

Additionally, it would facilitate scientific investigation by enabling measurement-taking by researchers utilising its radio and infrared optical communication links, and its antennas would be useful in radio astronomy.

Yew thinks that there are some areas and circumstances where his approach would likely be more applicable and serve as a necessary alternative to LunaNet, particularly in situations where network disruptions are brought on by power or signal outages, even though LunaNet’s potential is impressive, it is still a long way from being operational.

According to him, “When we’re doing human expeditions, you always want [backup systems] for very dangerous missions,” he said in an interview with PopSci, adding that his AI would not be “tied to the internet, per se,” although it can be.

Yew also cited the example of lunar crater exploration in the NASA press release, where the horizon would consist solely of the crater’s rim. Other geographical features on the Moon, like as ridges and boulders, may be difficult for a person to locate with accuracy but are nonetheless navigable by his system.

How lunar GPS would work

Yew is teaching an AI how to replicate lunar horizon characteristics as digital panoramas as part of his work on the lunar GPS.

This is done by using information from the Lunar Orbiter Laser Altimeter (LOLA) on NASA’s Lunar Reconnaissance Orbiter, as well as GIANT (Goddard Image Analysis and Navigation Tool) to confirm navigational data.

GIANT can analyse photos and calculate distances between visible landmarks with great speed and accuracy.

Yew outlined his method to PopSci, which makes use of a map’s outline to identify the location of the lunar surface in relation to space.

When the technology is fully developed, one use case for it will be to programme handheld devices with localised terrain and elevation data for lunar explorers.

Dr. Alvin Yew was contacted by Interesting Engineering (IE) to learn more about his research.Here are his exact words.

The following conversation has been minimally edited for flow and clarity.

Interesting Engineering: How is the AI trained and utilized to steer on the lunar surface? 

Dr. Alvin Yew: Even though everyone has been quite enthusiastic about AI as a technology, we’re still not there yet. The majority of our work to date has been focused on developing algorithms and conducting sensitivity analyses. Pretty dull material, but it prepares the way for the future.

 IE: What are the potential applications of this technology on Earth? 

One of the technological uses for Earth is localization without GPS. For instance, if you got lost while trekking and had no signal, or if your GPS is zigzagging about the city because of interference from nearby structures.

 IE: What is next for this research? How will it be tested?

A good query. The project’s further steps involve assessing the algorithms’ robustness, selecting the hardware requirements/limitations, deciding on limits for use cases, doing field testing using lunar analogues, and finally performing a demonstration on the moon.

I respond “good question” to the last bullet because my team and I have too many applications. Thus, it is likely that this will advance slowly during the coming year! I currently devote the majority of my time to collecting and enriching planetary and lunar samples in-situ in order to search for trace species, such as noble gases, hydrocarbons, and organics.For example, the In-Situ Resource Utilisation programme focuses on finding ways to make things from elements that are naturally occurring in space and close to their surroundings. This programme needs to be developed in order to set up a long-lasting infrastructure for when NASA returns to the Moon via the Artemis programme and later space excursions.

Categorized in: