An innovative new drone dubbed the “Phantom Twist” can seemingly disappear from view, in a major new robotics advancement offering a close-to-invisible aerial platform.
Its name describes the drone’s rapid spinning motion, which rotates 25 times per second—too fast for the human eye to resolve—thereby acting as a sort of cloak shielding it from visible detection.
The Northwestern University engineering team behind the Phantom Twist revealed their drone during a recent talk at the Sydney, Australia, Robotics: Science and Systems 2026 conference on July 16.
Stealth Drones
“Most efforts to hide drones focus on making them look like their surroundings,” said Northwestern’s Michael Rubenstein, who led the work. “Instead, we asked whether we could design the drone itself around the way humans perceive motion. This idea of low visibility through persistent motion is something few people have explored.”
Hiding drones is not a new idea. Other engineers have attempted a variety of techniques, including camouflage, transparent materials, and light-bending to obscure the presence of drones and robots. However, all of those focused on the material the drone’s exterior was constructed from, allowing it to blend into the background. The Northwestern team’s innovation was in how the object moves.
To our eyes, fan blades disappear because they are moving so fast that our eyes cannot track them. The “motion blur” created at these speeds allows the Phantom Twist to almost disappear, leaving only a faint apparition perceptible to the eye.
Nature Drones
While several military applications are immediately obvious, the team instead says their work could be useful in developing noninvasive wildlife and environmental monitoring. Loud drones can scare wildlife, disrupting them as they scatter, but by blending in, the Phantom Twist could capture a much more accurate picture of natural behaviors.
Their perception-based design eschews the traditional quadcopter design for a single propeller, with a body that spins in the opposite direction, creating the optical illusion that the drone essentially disappears.
“For a typical quadrotor drone, the propellers are spinning, but the robot is stationary,” Rubenstein said. “So, you still see its body. For our drone, the whole thing is rotating, so there are no stationary parts.”
Creating the Phantom Twist
Using AI and optimization algorithms, the Northwestern team began by generating 20,000 different arrangements of a drone’s propeller, motor, circuit board, and other necessary components. Eventually, the algorithm landed on the most promising configurations, after which the researchers simulated and superimposed them onto real-world backgrounds.
Another computer model mimicked human perception, allowing the engineers to assign a perception score for how visible each design would be to human eyes. An algorithm then optimized those designs by moving components to minimize the visibility score even further.
“The design process was fully automated,” Rubenstein said. “Then, when we were confident that a drone met all our criteria, we built it.”
A Revised Drone
The key feature of this design is that the individual components are at such varying heights and angles that they do not overlap as it spins, maximizing the motion-blur effect and creating the appearance of a faint cloud rather than a solid object.
“The human eye takes time to accumulate signals, roughly analogous to the exposure time of a camera,” said Alexander, an expert in computer vision. “When an object spins quickly, we perceive it as blurring out and losing distinct features. Because this new drone is almost entirely transparent, its few opaque components are visually averaged with the background for an overall appearance of a slight haze.”
Currently impeding the drone from being even more imperceptible to the eye are the wires and support rods that remain visible, as well as the engine noise it produces, which is readily discernible. Going forward, the team plans to continue improving Phantom Twist in future iterations to incorporate more transparent materials and reduce the noise it produces.
Ryan Whalen covers science and technology for The Debrief. He holds an MA in History and a Master of Library and Information Science with a certificate in Data Science. He can be contacted at ryan@thedebrief.org, and follow him on Twitter @mdntwvlf.