These Humanoid Robots Can Now Move Like Ronaldo and LeBron James

Think athletes are safe from being replaced by robots? Think again.

Researchers at Carnegie Mellon University and NVIDIA have developed a groundbreaking training technique that enables humanoid robots to perform complex athletic moves with unprecedented agility. From Cristiano Ronaldo’s signature mid-air spin celebration to Kobe Bryant’s iconic fadeaway jump shot, these robots are starting to look more like professional athletes than clunky machines.

At the heart of this innovation is a new framework called Aligning Simulation and Real Physics (ASAP), which bridges the gap between virtual simulations and real-world performance. This advancement marks a significant step toward robots that not only walk and talk like humans but also move like them.

How Do Robots Mimic Athletic Moves?

The ASAP framework tackles one of robotics’ biggest challenges: getting robots to perform agile, whole-body movements in the real world.

Here’s how it works:

• Pre-Training in Simulation: The robots are first trained in a virtual environment using real human motion data to develop motion-tracking algorithms.

• Real-World Deployment: These algorithms are then transferred to real robots, which adjust their movements based on the differences between simulated and actual physics.

This two-stage process allows robots to mimic complex athletic maneuvers, including:

• Cristiano Ronaldo’s “Siu” Celebration: A 180-degree mid-air spin that lands with flair.

• LeBron James’s “Silencer” Celebration: Single-leg balancing with precision.

• Kobe Bryant’s Fadeaway Jump Shot: A difficult leap that ends in a smooth, one-foot landing.

And it doesn’t stop there—the robots can also perform forward and side jumps over 1 meter, showcasing a level of dexterity rarely seen in humanoid machines.

What Makes These Robots So Agile?

While the robots may still appear a bit stiff compared to professional athletes, their agility is far beyond what most humanoid robots have achieved. This is largely due to ASAP’s delta action model, a correction mechanism that compensates for the inevitable discrepancies between simulation and real-world dynamics.

According to the researchers, this model reduced tracking errors by up to 52.7% compared to previous techniques. This breakthrough means the robots can handle complex movements that were previously impossible, bringing them closer to real human motion.

"Our approach significantly improves agility and whole-body coordination across various dynamic motions,” the researchers noted, emphasizing how ASAP is paving the way for versatile humanoid robots in real-world applications.

Why Is This Such a Big Deal?

Developing robots with this level of dexterity is one of the toughest challenges in robotics. Unlike simple locomotion—where robots are trained just to walk—athletic movements require constant, real-time adjustments. Humans naturally coordinate hundreds of joints, balancing forces and compensating for momentum shifts, but getting robots to do the same has been a monumental task.

To put this in perspective:

QWOP, a simple video game where players control an athlete’s four joints to make an athlete run, is notoriously difficult. Now imagine scaling that up to 21 articulations like ASAP handles—or even the 300+ joints in the human body.

The Race for Humanoid Robotics

The development of agile humanoid robots is heating up, with companies and academic institutions pouring resources into the field.

Some of the major players include:

• Tesla’s Optimus Project

• Figure AI’s recent humanoid robot announcements

• Boston Dynamics’ Atlas, famous for its parkour-like abilities

Meanwhile, universities like Stanford and the University of Bristol are refining methods to improve robotic agility and dexterity, pushing the boundaries of what these machines can achieve.

What’s Next? A Robot World Cup?

The researchers behind ASAP are already planning future developments to enhance their robots' abilities. Their focus includes:

• Damage-Aware Policy Architectures: To prevent hardware failures during complex movements.

• Markerless Pose Estimation: Reducing reliance on motion-capture systems by using onboard sensors.

• Improved Adaptation Techniques: Enhancing how robots adjust from simulations to real-world environments for even greater efficiency.

With these advancements, the idea of a robot soccer match or even an all-robot World Cup might not be as far-fetched as it sounds.

Looking Ahead

While robots aren’t about to replace Cristiano Ronaldo or LeBron James anytime soon, the leaps in agility and coordination made possible by ASAP mark a major milestone in humanoid robotics. As these technologies continue to develop, we might see robots not just mimicking humans but potentially surpassing them in certain athletic feats.

For now, it’s a fascinating glimpse into a future where the line between human and machine becomes even blurrier—on and off the field.

Editor’s Note: This article was created by Alicia Shapiro, CMO of AiNews.com, with writing, image, and idea-generation support from ChatGPT, an AI assistant. However, the final perspective and editorial choices are solely Alicia Shapiro’s. Special thanks to ChatGPT for assistance with research and editorial support in crafting this article.