The SONATE-2 mission will verify novel artificial intelligence (AI) hardware and software technologies

The SONATE-2 mission will verify novel artificial intelligence (AI) hardware and software technologies

There is a lot of talk about artificial intelligence at the moment, but in space travel, AI is still in its infancy. A German satellite in space is supposed to change that.

Germany’s space engineers could hardly have found a more musical name: SONATE is the name of their satellite. This name is also an abbreviation for SOlutus NAno Satellite – an unbound, free, independently operating mini-satellite.

Because that’s exactly what it’s about: SONATE-2 is designed to operate without human intervention and rely entirely on AI for its mission.”SONATE-2 is about the size of two shoe boxes,” explains Hakan Kayal, the head of the Interdisciplinary Center for Extraterrestrial Sciences at the University of Würzburg. The satellite has two fold-out solar panels and four deployable antennas.

What is water?

Visually, what the aerospace engineer describes doesn’t lookvery impressive. It’s the software and hardware that make SONATE-2 special.This includes eight cameras. “These cameras look towards the earth and record different regions that we have previously defined,” says Kayal .”We want to use these recordings to train the AI ​​​​on board.”

The scientists laid the foundation for this training on Earth before the launch: The SONATE-2 software was taught which landscape formations look like what. “What water is, what is not water, what reflections are and how snow differs from clouds – all of this has already been pre-trained.” Oleksii Balagurin from the University of Würzburg’s aerospace informatics department was responsible for this. He is the projectmanager of SONATE-2. “We want to use our AI to distinguish between earth, water and clouds, for example.”

The AI ​​of SONATE-2 can now do that. Now it’s off into Earth orbit. There, it will apply the knowledge it has learned and show what it has learned. “The goal is to detect anomalies,” says scientist Kayal. To do this, the AI ​​​​has learned what the Earth looks like. “If its cameras discover something that the AI ​​​​doesn’t yet know, it will be detected as an anomaly.”

In search of anomalies

The satellite’s cameras look down and compare what they see on the ground with what they have learned on the ground. If something doesn’t match, SONATE-2 will pay attention to certain objects. These could be, for example, circular irrigation devices, ie systems with geometric shapes.

“We taught the AI ​​what a desert is – and if a round irrigation system appears in it, the system should be able to recognize it as an anomaly,” explains Kayal. Anomalies could also be an oasis in the middle of the savannah or cracks in an ice sheet.

The Federal Ministry of Economics is funding SONATE-2 with 2.6 million euros. The plans are even more ambitious: in the future, this AI in space will be expanded to other planets or moons in the solar system. In a next step, SONATE-2 will turn its cameras away from the Earth and instead look out into the solar system. Because who knows what kind of anomalies are there -whether circles, triangles or lines. These include formations that have arisen due to geological activities, just like on Earth. But biological or biochemical activities can also produce geometric shapes.

Are we alone in space?

Ultimately, the next generation of SONATE satellites could even help answer the question of whether we are alone in space – or at least in the solar system. “It is conceivable that artefacts will be discovered in the solar system that are not of human origin,” believes Kayal. “It may be that alien spacecraft flew past a long time ago, perhaps landed or crashed, or parts of them could be present in the solar system.”

And then they should still be there, so the thinking goes.”It could be that with the technology we are now testing, such potentially artificial artefacts can also be recognized.” Because for AI, extraterrestrial technology would not be unusual ; it would just be another anomaly.

SONATE-2 successfully launched

On Monday, SONATE-2’s journey began on board a “Falcon9” rocket, and analyses are beginning on the ground. Project leader Balagurin and his team will receive the data from space on the Hubland campus of the University of Würzburg. “We are in the hot phase in which we simulate SONATE-2 flying over Germany.”

The satellite will be accessible for ten minutes three times a day. “In these ten minutes, we have to upload our daily schedule and download the data from experiments.” Then it will soon become clear what the extraterrestrial AI can do.

Artificial Intelligence in Space Exploration

Exploring space has always exhibited human curiosity and inventiveness. From mankind’s first lunar walk to the endeavors of Mars rovers, the human pursuit to investigate the universe keeps progressing. In recent times, artificial intelligence (AI) has become a monumental force in this field, transforming how we comprehend and explore the immense expanses of space.

AI’s role in space exploration has ignited a new era of effectiveness, creativity, and revelation. Its uses encompass independent steering and data examination to spacecraft upkeep and planetary investigation.

Self-sufficient Navigation and Operations

One of the key functions of AI in space exploration is self-sufficient navigation. Spacecraft and rovers integrated with AI can steer and make judgments without constant human involvement. This independent functionality is crucial for missions to far-off planets or moons, where communication lags can extend from minutes to hours.

For example, AI algorithms are utilized by NASA’s Mars rovers such as Curiosity and Perseverance to scrutinize terrain, devise paths, and evade barriers. This capability enables them to explore with greater efficiency and safety, covering more ground and carrying out more scientific experiments compared to direct human control.

Data Analysis and Understanding

Space missions produce substantial volumes of data, ranging from high-detail images to sensor readings and scientific metrics. AI excels at processing and interpreting extensive datasets, recognizing patterns, and deriving meaningful conclusions.

AI-powered tools can scrutinize data from telescopes, satellites, and rovers to pinpoint celestial bodies, discover irregularities, and even predict astronomical occurrences. For instance, the Kepler Space Telescope used AI to uncover numerous exoplanets by analyzing light patterns from distant stars, detecting potential planets through subtle luminosity variations.

Spacecraft Upkeep and Repair

AI holds a crucial role in preserving and mending spacecraft, particularly during extended missions. Anticipatory maintenance algorithms can supervise the condition of spacecraft systems, foresee possible malfunctions, and recommend preventive actions. This capability is essential for ensuring the durability and reliability of space missions.

Robotic systems outfitted with AI can also execute repairs in space. For instance, the Robonaut, a humanoid robot developed by NASA, can perform tasks that would be formidable or hazardous for astronauts, such as repairs on the International Space Station (ISS).

Planetary Exploration

AI enriches planetary exploration by enabling more advanced and autonomous scientific inquiries. AI-driven instruments can analyze soil samples, detect chemical compositions, and identify indications of life or habitable environments.

For example, the AI-based tool AEGIS (Autonomous Exploration for Gathering Increased Science) on NASA’s Curiosity rover can independently select and scrutinize rock targets, giving priority to those that are most scientifically intriguing. This automony boosts the efficiency and scientific output of the mission.

How NASA Utilizes AI in Space Exploration

NASA, the trailblazer of space exploration, is persistently striving to address these profound questions. In recent times, Artificial Intelligence (AI) and Machine Learning (ML) have emerged as vital tools in NASA’s quest to explore and comprehend the universe. These advanced technologies not only amplify our ability to investigate space but also overhaul the way we analyze vast data troves, make crucial decisions, and conduct scientific investigations in the most extreme environments acknowledged by humankind.

The Role of AI and Machine Learning in NASA’s Missions

The integration of AI and ML at NASA is revolutionizing space exploration, enabling more efficient operations, deeper scientific insights, and groundbreaking discoveries. Here’s how NASA employs these cutting-edge technologies:

1. Self-Driving Rovers on Mars

Spirit, Opportunity, and Curiosity Rovers

Even before companies like Tesla and Google popularized self-driving cars, NASA was spearheading self-directing technology for Mars rovers. The Spirit and Opportunity rovers, which landed on Mars in 2004, were equipped with a Machine Learning navigation system called AutoNav. This system enabled the rovers to autonomously navigate the rugged Martian terrain, sidestepping obstacles such as rocks and sand dunes.

Curiosity, which landed in 2012, continues to employ and enhance this technology. It utilizes AutoNav and the AEGIS (Autonomous Exploration for Gathering Increased Science) algorithm to spot intriguing rock formations. As communication with Earth is limited, AEGIS aids Curiosity in prioritizing and relaying the most scientifically significant images.

2. As astronauts set out on longer journeys beyond Earth’s orbit, maintaining their well-being becomes increasingly important. NASA’s Exploration Medical Capability (ExMC) project utilizes ML to create independent healthcare solutions customized to astronauts’ requirements. These solutions are designed to adapt to astronauts’ needs, providing immediate medical aid in space where direct communication with Earth-based doctors is not feasible.

3. The exploration of exoplanets—planets outside our solar system—is a major focus for NASA. The Planetary Spectrum Generator uses ML to construct intricate models of these planets’ atmospheres. By examining spectral data, ML algorithms can forecast the existence of elements such as water and methane, which are signs of potential life. This technology empowers NASA to uncover and investigate new planets, bringing us closer to addressing the enduring question of whether we are alone in the universe.

4. Robonaut, NASA’s robotic astronaut, is engineered to support human astronauts in tasks that are perilous or tedious. Fitted with advanced sensors and AI, Robonaut can independently carry out various functions. Machine Learning enables Robonaut to learn and adjust to new tasks, making it an invaluable companion in space exploration and enhancing NASA’s research capabilities.

Robonaut also possesses numerous advantages over human personnel, including advanced sensors, exceptional speed, compact design, and significantly greater flexibility. The development of Robonaut involved the utilization of advanced technology, such as touch sensors at its fingertips, a wide neck travel range, a high-resolution camera, Infra-Red systems, advanced finger and thumb movement, and more.

5. Getting lost on Earth is not a major issue, thanks to GPS. However, what if you were to get lost on the Moon? GPS does not function there! Nonetheless, NASA’s Frontier Development Lab is working on a project to provide navigation on the Moon and other celestial bodies without relying on multiple costly satellites.

This innovative solution involves utilizing a Machine Learning system trained with 2.4 million images of the Moon held by NASA. By creating a virtual lunar map using neural networks, the system allows for precise navigation. If you become lost on the Moon, you can capture images of your surroundings, and the Machine Learning system will compare these images with its extensive database to determine your location.

Despite not yet being flawless, this method significantly exceeds existing navigation techniques and can be adapted for other planetary surfaces as well. NASA is optimistic that this technology can also be employed on Mars, providing crucial navigation support for future explorers on the Red Planet.

6. NASA is employing AI to develop mission hardware. AI-designed components, resembling organic structures, are lighter, stronger, and faster to develop compared to traditional designs. This innovation not only enhances the performance and reliability of spacecraft but also accelerates the development process, allowing for quicker mission readiness (NASA).

NASA is integrating generative AI into space. The organization recently revealed a series of spacecraft and mission hardware designed using the same type of artificial intelligence that generates images, text, and music from human prompts. Known as Evolved Structures, these specialized parts are being incorporated into equipment including astrophysics balloon observatories, Earth-atmosphere scanners, planetary instruments, and space telescopes.

7. AI plays a crucial role in SpaceX’s rocket landings by enabling independent navigation and control, processing real-time sensor data, and utilizing machine learning for predictive analytics. It computes optimal landing trajectories, ensures accuracy, and integrates with ground systems for real-time adjustments. AI-driven systems also provide redundancy for fault tolerance, significantly boosting landing reliability and success rates. This technology has enabled SpaceX to successfully recycle rockets, reducing space travel costs.

Future of AI in Space Exploration

Artificial Intelligence is positioned to transform space exploration, unlocking new opportunities and reshaping our comprehension of the universe. For example, NASA’s Parker Solar Probe, set to reach the Sun’s outer atmosphere in December 2024, will utilize advanced AI systems to withstand extreme temperatures of up to 2500℉ (1370℃) and collect crucial data with its magnetometer and imaging spectrometer. This mission aims to enhance our understanding of solar storms and their impact on Earth’s communication technologies.

AI’s role extends beyond this, as it will significantly improve the monitoring of Earth-orbiting satellites and manage spacecraft on extended missions. By integrating AI with robotics, future missions may deploy autonomous robots capable of exploring distances and environments beyond the reach of human astronauts.

Artificial intelligence (AI) is revolutionizing many industries, and space exploration is no different. As we journey deeper into space, AI becomes increasingly crucial in tackling the challenges of extended communication delays, managing massive data sets, and enabling autonomous robotic planetary exploration systems.

Handling Enormous Data Amounts

The significant increase in space data collected from satellites, telescopes, and interplanetary probes necessitates the analytical capabilities of AI. Today’s space instruments produce terabytes of data daily, far exceeding what scientists can manually review.

AI automation assists in categorizing and processing continuous streams of images, sensor readings, and spectral data. For instance, AI techniques are utilized in NASA’s Mars Reconnaissance Orbiter to filter and prioritize over six megabits per second of data. Scientists trained these AI algorithms to identify key features from billions of images of Mars’ surface.

Additionally, astronomers use AI to sift through astronomical data sets. Neural networks have been trained to detect exoplanets from fluctuations in light curves captured by the Kepler space telescope. These AI tools also classify galaxy types and group stars based on shared motion.

NASA and Google collaborated 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 missed. This success prompted the utilization of AI in analyzing data from NASA’s TESS mission to identify potential exoplanets.

“New methods of data analysis, such as this initial research to implement machine learning algorithms, promise to continue yielding significant advancements in our understanding of planetary systems around other stars. I’m confident there are more groundbreaking discoveries waiting to be unearthed in the data.” Jessie Dotson, NASA Ames Research Center’s Kepler project scientist.

In a study published in Astronomy and Astrophysics, led by University of Leeds’ researcher Miguel Vioque, AI was incorporated in the data analysis of the Gaia space telescope, leading to the identification 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 great potential for automating spectral data analysis from future missions to locations like Saturn’s moon Enceladus, where rapid onboard processing will be crucial for identifying potential signs of microbial extraterrestrial life in ice plumes emanating from a subsurface ocean.

Enabling Autonomous Robotic Planetary Exploration

AI provides advanced autonomy to robotic rovers on planetary surfaces like Mars, empowering them with capabilities for vision-based navigation, path planning, object detection, and adaptive mission prioritization, allowing them to traverse challenging and unfamiliar terrain using onboard maps and sensor data.

For instance, NASA’s Curiosity and Perseverance rovers leverage AEGIS, a powerful AI system, to create autonomous 3D terrain maps and identify rock features and soil composition. It can even suggest the day’s activities based on terrain complexity, energy usage, and scientific value.

Such intelligent capabilities will become increasingly crucial as future rover missions target more distant destinations with greater communication delays from Earth, such as gas planets and their icy moons. Additionally, AI enables autonomous navigation and adaptable scientific exploration; rovers can respond to discoveries immediately rather than waiting for delayed commands.

AI also aids in entry, descent, and landing (EDL) – the riskiest phase for 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 accurately reach designated landing zones. As agencies plan more ambitious robotic missions, AI provides the advanced autonomy to explore harsh and unfamiliar environments.

Supporting Astronaut Health

The mental and physical strain during multi-year missions creates a need for improved astronaut medical care. AI holds promise for enhancing 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. Holistically combining real-time vital signs, behavioral indicators, and environmental conditions allows for sophisticated diagnostics, early risk alerts, and personalized treatment plans.

For instance, the Crew Interactive Mobile Companion (CIMON), developed by Airbus, IBM, and the German Aerospace Center, is an AI robot controlled by voice that traveled to the International Space Station (ISS) in 2018.

CIMON can see, hear, understand, and speak using voice and facial recognition, enabling it to move around the space station, locate and retrieve items, document experiments, and display procedures.

CIMON’s primary function is to serve as a comforting and empathetic companion that can detect levels of stress. It has been trained to provide psychological support using Watson’s natural language abilities and can guide astronauts through therapeutic exercises to improve their mood.

Further advanced systems on the ISS and lunar Gateway will be tested to predict the needs of astronauts, offer suggestions, and automate routine tasks. Future Mars missions, which face communication delays with ground control, will also utilize AI virtual assistants for psychological support.

In conclusion, AI plays a transformative role in space exploration by analyzing extensive data from celestial bodies and forecasting potential hazards such as solar storms and space debris. It enhances spacecraft autonomy, reduces human dependency, and supports astronauts in operations, navigation, and satellite monitoring.

Artificial intelligence (AI) and robotics are accelerating human problem-solving. AI represented a significant improvement over traditional computing, as it lacked data backups and recovery options.

The advancements in AI have made it valuable across a multitude of scientific domains. From robotics in packaging to machine learning, AI is contributing to progress in various fields.

The benefits of AI aren’t restricted to applications on Earth. Here are some examples of how AI is advancing current space endeavors:

  • Assisting with mission design and planning
  • AI is simplifying the planning of missions beyond Earth for mission designers.
  • New space missions build upon knowledge gained from previous studies. Limited data can present challenges for current scientists when planning missions.
  • AI addresses this issue by providing authorized individuals with access to all space missions. With AI, mission designers can easily access relevant data.
  • One example of such a solution is Daphne, an intelligent assistant for creating Earth observation satellite systems.
  • Systems engineers on satellite design teams use Daphne to access data, feedback, and answers to mission-related questions.
  • Aiding in the manufacturing of satellites and spacecraft
  • Engineers fabricate intricate satellites and spacecraft using costly equipment.

The manufacturing process involves intricate and repetitive tasks that require precision. Engineers often require specialized facilities to fabricate satellites and spacecraft to prevent potential contamination.

This is where AI-enabled systems and robotics come into play. Scientists use AI and robots to alleviate their workload, allowing humans to focus on tasks that computers cannot perform.

AI can accelerate the assembly of satellites. AI-enabled systems can also analyze the process to identify areas for improvement.

Scientists also utilize AI to review the work and ensure its accuracy.

Cobots, or collaborative robots, also contribute to satellite and spacecraft development. These cobots interact with humans within a shared workspace.

They help reduce the need for human labor in clean rooms. They carry out reliable manufacturing tasks and minimize human error.

Aiding in the processing of satellite data

Earth observation satellites generate vast amounts of data. Ground stations receive this data in intervals over time.

Artificial intelligence can support this effort by conducting detailed analysis of satellite data. AI is an effective tool for analyzing big data.

Scientists use AI to estimate heat storage in specific areas and to calculate wind speed by combining meteorological data with satellite imagery.

It can also estimate solar radiation using geostationary satellite data.

Assisting with navigation systems

On Earth, individuals rely on navigation systems like GPS for tools such as Google Maps. Currently, there are no equivalent navigation systems in space.

However, scientists can utilize imagery from observation satellites. One such satellite is the Lunar Reconnaissance Orbiter (LRO), which provides data to support future lunar missions.

In 2018, NASA and Intel utilized LRO data to develop an intelligent navigation system. The system used AI to generate a map of the moon.

Monitoring the health of satellites

Operating satellites involves complex processes. Equipment malfunctions and satellite collisions can occur at any time.

To address this, satellite operators utilize AI to monitor satellite health. AI-enabled systems can check sensors and equipment and alert scientists when attention is needed.

In some cases, AI-enabled systems can even take corrective actions.

Scientists use AI to control the navigation of satellites and other space assets. AI uses past data to recognize satellite patterns and can alter the craft’s trajectory to prevent collisions.

AI can also support communication between Earth and space.

This form of communication can be challenging due to interference, which may arise from other signals or environmental factors.

Thankfully, AI has the capability to manage satellite communication in order to tackle potential transmission issues. AI-powered systems can calculate the necessary power for transmitting data back to Earth.

Improves satellite pictures

Multiple images are generated by satellites every minute. Each day, they handle vast amounts of data.

This data includes weather and environmental images. Additionally, these satellites capture Earth images, which presents numerous challenges.

AI aids in interpreting, analyzing, and comprehending satellite images. With the help of AI, humans can review the millions of images produced by space assets.

AI can analyze satellite images in real time. It can also detect any issues with the images if they exist.

One advantage of utilizing AI is that, unlike humans, AI does not require breaks. This means AI can process more data more quickly.

Employing AI for this purpose eliminates the need for extensive communication to and from Earth. This can decrease processing power and battery consumption while streamlining image capture.

These are the ways in which AI is progressing space exploration efforts.

This demonstrates that AI not only enhances the quality of life on Earth but also enables space exploration.

It also demonstrates that the various benefits of AI in space make venturing into the unknown safer.

Space exploration is one of humanity’s most challenging and thrilling pursuits. It necessitates the integration of scientific knowledge, technological innovation, and human bravery.

However, there are numerous limitations and risks associated with sending humans and spacecraft into the vast and unexplored realms of the cosmos. This is why artificial intelligence (AI) is crucial in discovering new worlds and broadening our horizons.

AI is the field of computer science that involves creating machines and systems capable of performing tasks that typically require human intelligence, such as reasoning, learning, decision-making, and problem-solving. AI can help us overcome some of the challenges and improve certain space exploration opportunities. Here are seven remarkable applications of AI in space exploration:

Assisting Astronauts

AI can aid astronauts in performing various tasks on board the spacecraft or the space station, such as monitoring systems, controlling devices, conducting experiments, or providing companionship. For example, CIMON is an AI assistant that can interact with astronauts on the International Space Station (ISS) using voice and facial recognition. CIMON can assist astronauts with procedures, answer questions, or play music. Another example is Robonaut, a humanoid robot that can work alongside or instead of astronauts in hazardous or routine missions.

Designing and Planning Missions

AI can assist in designing and planning space missions more efficiently and effectively by utilizing extensive data from prior missions and simulations. AI can also optimize mission parameters, such as launch date, trajectory, payload, and budget. For instance, ESA has developed an AI system named MELIES that can aid mission analysts in designing interplanetary trajectories using genetic algorithms.

Spacecraft Autonomy

AI can empower spacecraft to function autonomously without depending on human intervention or communication from Earth. This is particularly beneficial for deep space missions, where communication delays can be significant. AI can assist spacecraft in navigation, avoiding obstacles, adapting to changing environments, or responding to emergencies. For example, NASA’s Mars 2020 rover Perseverance uses an AI system called Terrain-Relative Navigation to analyze images of the Martian surface and adjust its landing position accordingly.

Data Analysis

AI can help analyze the vast amounts of data collected by space missions, such as images, signals, spectra, or telemetry. AI can process data faster and more accurately than humans, uncovering patterns or anomalies that humans might overlook. For instance, NASA’s Kepler space telescope employed an AI system based on neural networks to discover new exoplanets by detecting their transit signals.

Space Communication

AI can improve communication between spacecraft and Earth or between spacecraft. AI can optimize communication bandwidth, frequency, power, or modulation. AI can also enhance the security and reliability of communication links by identifying and correcting errors or interference. For example, NASA’s Deep Space Network utilizes an AI system called Deep Space Network Now that can monitor and predict the status and availability of the communication antennas.

Space Debris Removal

AI can help mitigate the issue of space debris, which consists of defunct or abandoned objects orbiting Earth and posing a threat to operational spacecraft. AI can aid in tracking and cataloging space debris using radar or optical data. AI can also assist in designing and managing missions to remove or deorbit space debris using robotic arms or nets—for example, ESA’s e.The deorbit mission plans to utilize an AI system that can autonomously capture a derelict satellite using a robotic arm.

NASA’s Dragonfly mission plans to use an AI system to search for signs of life beyond Earth. AI can help identify habitable planets or moons by analyzing their physical and chemical characteristics. It can also use biosignatures or biomarkers to search for signs of living organisms or their products. By using spectroscopy or microscopy techniques, AI can detect possible life forms. For instance, the mission aims to fly a drone-like rotorcraft on Saturn’s moon Titan and collect samples for signs of prebiotic chemistry.

Suddenly, circular openings appeared on the surface of Mars that hadn’t been present before. In photographs of Saturn’s moon Enceladus, geysers were found that shoot powerful jets of steam into space. Additionally, images transmitted to Earth by the Mars rover Curiosity revealed formations resembling fossilized worms.

All of these occurrences, some of which seem temporary, were discovered either by chance or because humans spent considerable time analyzing images from Earth’s neighboring planets. “Artificial intelligence technologies would significantly simplify the identification of previously unrecognized anomalies,” states Hakan Kayal, a Professor of Space Technology at Julius-Maximilians-Universität (JMU) Würzburg in Bavaria, Germany.

Science is still in the early stages

Can artificial intelligence (AI) be utilized in astronautics? According to Professor Kayal, research in this area is still in its early phases: “Only a few projects are currently in progress.”

For an AI to identify unknown occurrences, it must be initially trained. It needs to be “fed” known information so that it can learn to recognize the unknown. “There are already satellites operated with AI that are trained on Earth before being sent into orbit. However, we have different plans: We intend to train the AI aboard a small satellite under space conditions,” explains the JMU professor.

This endeavor is challenging but attainable: “Miniaturized IT systems are becoming increasingly powerful. We are allowing sufficient time for AI training, which means the learning process in orbit can span several days.”

Interplanetary missions as a long-term objective

But why move the training of the AI to space, to miniature computers? Wouldn’t it be simpler to implement this with mainframe computers on Earth? Hakan Kayal has a clear vision for the future. He aims to use small satellites equipped with AI not just for monitoring Earth but also for interplanetary missions to uncover new extraterrestrial phenomena, possibly even evidence of extraterrestrial intelligences.

“As soon as interplanetary travel begins, communication with the satellite faces limitations,” states the professor. As the distance from Earth increases, data transfer times lengthen; “you cannot continue to send data back and forth. That’s why the AI needs to learn autonomously on the satellite and report only significant discoveries back to Earth.”

Launch into orbit anticipated in 2024

Kayal’s team, led by project leader Oleksii Balagurin, plans to implement this technology on the small satellite SONATE-2 and assess its performance in orbit. The Federal Ministry for Economic Affairs and Energy is supporting the project with funding of 2.6 million euros. The initiative commenced on March 1, 2021, with the satellite scheduled for launch into orbit in spring 2024. The mission’s duration is expected to be one year.

The small satellite from Würzburg will be approximately the size of a shoebox (30x20x10 centimeters). Its cameras will capture images in various spectral ranges while monitoring the Earth. The image data will be processed by the onboard AI, which will automatically identify and categorize objects. The technology will undergo thorough testing around Earth before it potentially embarks on an interplanetary mission in the future. Hakan Kayal has already included this prospective mission, named SONATE-X, in his research agenda—the X stands for extraterrestrial.

Students can get involved

SONATE-2 will feature other innovative and highly autonomous capabilities. In comparison to its predecessor, SONATE, the sensor data processing system will be further miniaturized and optimized for energy efficiency. Furthermore, new types of satellite bus components, including advanced star sensors for self-governing attitude control, will be implemented. The cameras will not only capture and document static objects but also brief, transient events like lightning strikes or meteors.

The team working on SONATE-2 will consist of around ten members. Students are also encouraged to participate—either as assistants or through bachelor’s and master’s thesis projects. Educating the next generation in this innovative technology is integral to the project. In addition to its computer science programs, JMU offers both Bachelor’s and Master’s degrees in Aerospace Informatics along with a Master’s program in Satellite Technology.

The SONATE-2 project is funded by the German Aerospace Center (DLR) using resources from the Federal Ministry for Economic Affairs and Energy (BMWi) based on a resolution from the German Bundestag (FKZ 50RU2100).

FredMT Admin Avatar

Leave a Reply

Your email address will not be published. Required fields are marked *