A normal ballpoint pen won’t write in area. Without gravity, the ink refuses to circulate. This easy failure illustrates a profound headache in area exploration: instruments designed for terrestrial use typically grow to be ineffective in a microgravity surroundings. Robots, for all their technological sophistication, aren’t any exception.
Autonomous free-flying robots aboard the International Space Station (ISS) regularly lose their bearings. Without gravity to tell apart up from down, even precision sensors undergo from accumulating errors, inflicting the machines to float. Until not too long ago, astronauts typically needed to intervene manually, interrupting their tightly scheduled work.
The National Aeronautics and Space Administration (NASA) has discovered an answer to this persistent drawback by way of a collaboration with Professor Pyojin Kim and his staff on the Gwangju Institute of Science and Technology (GIST). An skilled in navigation know-how, the science of enabling robots to find out their 3D place and orientation, Professor Kim has proposed an algorithm to considerably suppress these errors. By lowering the ’absolute rotation error’ to inside about 1–2 levels on common, the staff has enabled robots to carry out long-term missions with out requiring human intervention.
We spoke with Professor Kim to debate how he tailored know-how for the cosmos and the breakthrough that retains NASA’s robots on observe.
Saving area robots with digital twin navigation
The International Space Station is a colossal orbital laboratory, roughly the dimensions of a soccer subject. It was constructed by connecting modules that had been developed by completely different nations. Inside the Japanese Experiment Module ’Kibo’, a free-flying NASA robotic named Astrobee is tough at work. Its mission is to take over routine chores, liberating astronauts to focus on analysis. With days scheduled to the minute, any time spent on upkeep is a expensive distraction for the crew.
In precise operation, nevertheless, Astrobee didn’t work as flawlessly as anticipated. It regularly misplaced its bearings, requiring astronauts to step in for recalibration. NASA engineers and Professor Kim’s staff collaborated to discover a method for the robotic to function reliably with out supervision, so the astronauts might give attention to their essential analysis.
The root of the disorientation is the absence of distinct gravity. Terrestrial robots depend on an Inertial Measurement Unit (IMU) to sense tilt and orientation relative to the gravity vector. Professor Kim factors out that “Terrestrial navigation algorithms are designed based mostly on gravity, making them troublesome to use instantly in area the place reference factors are lacking.“ As a outcome, tiny errors compound over time inflicting the robotic to fully lose its sense of course.
To counter this, the staff turned to Visual-Based Navigation (VBN), enabling the robotic to infer its orientation by seeing its environment by way of cameras. At first, the staff presumed that merely adopting established know-how can be enough. They had been improper.
The station’s inside is a chaotic jumble of cables, experimental rigs, and floating private objects. A view obtainable one minute may be blocked by a drifting pill the subsequent. This unpredictability confounded customary navigation methods. “We thought we might apply Earth-based know-how,“ recollects Professor Kim. “It didn’t carry out reliably in the ISS environments.“
The breakthrough got here in the type of ’digital twins’, exact 3D replicas of the bodily area. Using NASA’s blueprints, the staff constructed a sanitized digital mannequin of the ISS, stripped of all transient muddle. The robotic was programmed to cross-reference the messy real-time footage from its cameras with the pristine photos generated from the digital twin.
Professor Kim explains, “The digital twin serves as a floor fact, enabling the robotic to filter out visible noise and recalibrate its place.“
With this corrected knowledge, the robotic interprets its surroundings as a set of traces and planes. These extracted geometric options function a ’visible compass,’ offering an absolute directional reference. The system leverages the ’Manhattan World Assumption’, a precept positing that man-made environments consist primarily of orthogonal surfaces similar to partitions and flooring assembly at proper angles. The boxy modules of the ISS are a super testbed for this method. By locking onto these structural geometries, the robotic can triangulate its place with minimal error.
The staff achieved a ’drift-free’ navigation functionality. Upon making use of the brand new know-how, the common rotational error was diminished to 1.43 levels—a determine that doesn’t enhance over time. The robotic not requires a human hand to information it.
Professor Kim anticipates that this know-how can be helpful on Earth, not simply in area. It might function a information for drones and robots in indoor environments the place GPS indicators can not attain. The system depends on visible knowledge to detect structural patterns, making it very best for buildings stuffed with traces and planes. Professor Kim notes that “orientation methods based mostly on these structural options are relevant not solely to area stations but additionally to typical city settings.“
Insights from the NASA collaboration
Ask Professor Kim why humanity ought to enterprise into orbit, and his reply is refreshingly blunt: “Because area now holds actual financial and industrial worth, exhibiting business potential.“
With SpaceX proving that area could be a enterprise slightly than only a frontier, a wave of startups has emerged, focusing on the whole lot from lunar mining to satellite tv for pc meeting. Yet, NASA stays the silent companion behind this private-sector explosion. Its many years of gathered know-how and expertise type the bedrock upon which these new enterprises are constructed.
It was this ecosystem that drew Professor Kim, initially a drone specialist, into the fold. His journey started with an internship on the NASA Ames Research Center throughout his doctoral research. The middle was then in the thick of creating Astrobee. To mimic microgravity, researchers floated the robotic on air-bearing tables utilizing carbon dioxide jets, manipulating the lighting to carefully check its means to find itself.
This analysis was a pure match for Professor Kim’s experience. His time on the company revealed that terrestrial drones and area robots share the identical theoretical basis, regardless of their vastly completely different environments. The logic behind mapping an surroundings and figuring out location is common, differing in its software.
The connections made then have lasted almost a decade, evolving into the present joint analysis. Kim expressed his gratitude: “This analysis would have been unimaginable with out the assistance of my mentor on the time, Dr. Brian Coltin, my NASA colleagues, my present co-researcher Dr. Ryan Soussan, and Dr. Terry Fong, who supplied the alternatives for the internship and joint analysis.“
Professor Kim was notably struck by the company’s angle towards failure. During his time there, he witnessed NASA pursuing daring experiments, backed by substantial budgets and distinctive expertise. “Because solely profitable tasks are publicized, it seems as if they by no means fail,“ Professor Kim stated. “But behind each public triumph lie dozens of quiet failures.“ He notes the company’s energy lies in its willingness to endure these setbacks to realize a single breakthrough.
This give attention to actual affect formed their evaluation requirements as properly. Beyond typical educational metrics, NASA positioned specific emphasis on the real-world affect and sensible significance of the analysis. While it is not uncommon apply to submit two papers upon finishing a Ph.D, some researchers submitted just one, or opted to share their outcomes on preprint servers like arXiv slightly than in formal journals.
“Despite its conservative nature as a authorities company, NASA is surprisingly open in its method to analysis,“ Kim recalled. “I used to be impressed by the tradition of valuing the intrinsic worth and contribution of the analysis over mere outcomes.“
Sustained funding in science has paved the way in which for an unlimited industrial infrastructure and numerous area startups led by NASA alumni. Professor Kim factors to the sturdy U.S. ecosystem of producers specializing in ’space-grade’ elements able to withstanding excessive circumstances. It has created a virtuous cycle the place authorities funding nurtures expertise and know-how, fueling a wave of startups that drive the non-public sector.
For these aspiring to affix the company, Professor Kim provides recommendation grounded in realism.
“I need to offer you some practical recommendation. The researchers I met at NASA had been all from the world’s prime universities. It might sound cliché, however you have to excel at arithmetic and your research in normal. While it’s good to dream large, making that dream a actuality requires overwhelming competence. The door to the worldwide stage is all the time open. If you’re employed onerous to construct your abilities, the chance will certainly observe.“
This article was produced as a part of the NASA Impact Series by Popular Science Korea.
2025 PopSci Outdoor Gift Guide