AI digital twins aim to protect astronaut mobility on deep space missions

by Clarence Oxford
Los Angeles CA (SPX) Jan 29, 2026
West Virginia University researchers are developing artificial intelligence tools to help astronauts maintain movement control and overall physical health during long missions in microgravity environments such as space stations, the Moon and future voyages to Mars. Their work centers on building personalized computer models that capture how each astronaut moves and how their muscles are activated, allowing mission planners and crew members to better predict and prevent the degradation that occurs when gravity is removed for extended periods.

The project, led by Valeriya Gritsenko and Sergiy Yakovenko from the WVU School of Medicine and the WVU Rockefeller Neuroscience Institute, uses motion analysis and neuroscience to create a unique digital twin for each astronaut. These digital twins are designed to represent the individual relationships between limb movements and muscle activity, tracking how a person adapts to weightlessness and identifying early warning signs of trouble before they become mission-limiting problems.

In current operations, each astronaut depends on a large team on Earth to monitor vital signs and adjust exercise and countermeasures in response to health changes during flight. As human exploration pushes farther from Earth, communication delays and limited ground support will make that model harder to sustain, so the WVU researchers are working on more autonomous systems that can operate with the crew in deep space. Their goal is a noninvasive monitoring platform that follows astronauts through their daily routines and exercises, using observed movements to infer changes in neuromuscular function.

The team notes that even with extensive exercise programs, astronauts struggle to maintain proper coordination when they spend long periods in microgravity, which can leave them at risk of falling or losing balance when they attempt to walk after landing on another world or returning to Earth. The researchers warn that crews reaching Mars after many months in transit could experience severe disruption of coordination just when they need mobility the most. By quantifying how coordination changes over time, the digital twin system is intended to guide more effective countermeasures to preserve safe locomotion.

To construct their models, the scientists are collecting data from human volunteers who perform tasks in a controlled virtual reality environment, including reaching for objects and walking on a treadmill. Motion capture systems and wearable sensors record joint movements and measure the neural drive to muscles, while physics-based software similar to game engines simulates realistic body dynamics. By combining a person's size, weight and measured muscle actions with these simulations, the researchers can estimate the forces required for different movements and extrapolate how similar motions would unfold in orbit or in near weightlessness.

Once validated, the AI models are expected to monitor astronauts throughout mission phases, starting with structured exercise routines before launch, continuing through time in space and extending into rehabilitation after landing. By continuously analyzing movement patterns, the system can detect subtle changes that suggest emerging issues such as muscle loss, declining bone loading, balance problems or altered reflex responses. The model can then recommend adjustments such as increasing exercise frequency, changing loads or modifying routines to reduce the risk of serious impairment.

The researchers also plan for the models to help estimate how much astronauts have deconditioned while in orbit, providing input that supports individualized rehabilitation plans once they return to Earth. Such plans would target challenges like orthostatic intolerance, in which blood is not pushed to the head quickly enough during standing and can lead to loss of consciousness. By anticipating these responses, the system could help design safer reconditioning strategies that restore function while limiting the risk of falls and fainting.

The work is funded by a 750,000 dollar award from NASA through the NASA West Virginia Space Grant Consortium and the NASA West Virginia Established Program to Stimulate Competitive Research, or EPSCoR. Program leaders say the project advances NASA exploration goals while building capacity in the state for high technology industries, attracting federal investment and giving students hands-on experience at the intersection of aerospace, artificial intelligence and biomedical engineering.

WVU students from the School of Medicine and a range of fields including computer science and engineering are contributing to the research effort. The team includes graduate and undergraduate students who bring diverse skills in programming, physiology, robotics and data analysis, and the investigators emphasize that this mix of backgrounds is critical for progress on such a complex problem.

Beyond its applications in spaceflight, the digital twin approach could support terrestrial health care, especially telemedicine for people in rural areas or for patients who need remote monitoring of movement. By analyzing everyday motions, the models may help clinicians detect early signs of motor deficits, balance issues or neurodevelopmental delays that might otherwise go unnoticed without an in-person examination.

The researchers highlight parallels between the deconditioning that astronauts experience in microgravity and the loss of muscle and neuromotor control in people who are sedentary for long periods, including older adults or pregnant patients on extended bed rest. They see potential for the same tools that safeguard astronaut mobility on Mars-bound missions to assist vulnerable populations on Earth, linking space medicine research with broader public health benefits.