Purdue University students have just shattered the Guinness World Record for solving a Rubik’s Cube with their groundbreaking robot, achieving an astonishing time of just 0.103 seconds. This new record crushes the previous benchmark by nearly a third, but it’s not just about building a faster-moving robot. Their secret lies in a smart blend of high-speed vision systems, a specially customized cube, and advanced solving techniques borrowed from human speedcubers.
The race to build the fastest Rubik’s Cube-solving robot began back in 2014 with the creation of Cubestormer 3. Built with Lego Mindstorms and a Samsung Galaxy S4, it solved the puzzle in 3.253 seconds—an incredible feat at the time. Since then, engineers have relentlessly chipped away at the record, bringing it down to just milliseconds. Last year, Mitsubishi Electric in Japan claimed the record with a robot that solved the cube in 0.305 seconds. But Purdue’s team from the Elmore Family School of Electrical and Computer Engineering took it even further, setting a new standard with their robot, dubbed Purdubik’s Cube.
Reducing the solve time to under half a second meant moving beyond Lego components. The Purdue team opted for industrial-grade motors and precision hardware to maximize speed and accuracy. But the real game-changer was how they optimized the robot’s visual recognition system. Unlike human cubers, who can study the cube before starting, their robot had to quickly identify the colors of all sides in real-time. They employed high-speed Flir cameras, with a resolution of just 720×540 pixels, positioned to capture multiple sides simultaneously. These cameras, equipped with wide-angle lenses, could process images in as little as 10 microseconds, enabling rapid color detection.
To streamline data processing, Purdue’s team designed a custom image detection system that bypassed traditional image processing steps. It focused on tiny regions—only 128×124 pixels—reducing the data load and speeding up the identification process. Using RGB measurements from small sample areas on each square, their system determined colors faster than AI-based methods, prioritizing speed over absolute reliability—a trade-off that paid off in record time.
When it came to solving the cube, the team relied on existing software, specifically Elias Frantar’s Rob-Twophase algorithm, tailored for robotic manipulation. They also employed a clever technique known as corner cutting—starting to turn one side of the cube before completing the perpendicular turn on another—allowing overlapping movements that save precious milliseconds. High-speed footage reveals how this overlap helps the robot shave seconds off its solve time, but it also requires precise timing and careful control to prevent damage to the cube.
Speaking of the cube, Purdue’s team made modifications to ensure durability and optimal performance. They 3D-printed their cubes with stronger SLS nylon plastic, and applied special lubricants to keep the tension tight. This tensioning is critical; a tightly wound cube can be difficult to turn manually but enables smoother, faster rotations when driven by the robot’s motors. The cube’s internal mechanism was fine-tuned with custom metal shafts connected to high-speed servo motors, which can accelerate to an incredible 12 million degrees per second squared, then decelerate smoothly to ensure precise positioning.
Could Purdue’s robot break its own record in the future? Patrohay believes so, especially if they develop a more durable cube made from advanced materials like carbon fiber. Such a cube could withstand higher speeds, potentially driving the solve time even lower.
In essence, Purdue’s innovative approach combines cutting-edge hardware, clever software, and advanced manipulation techniques. Their achievement not only sets a new world record but also pushes the boundaries of what’s possible in robotics and speedcubing. As technology advances, the future of Rubik’s Cube solving looks faster and more exciting than ever.
