In a world where technology and sport intertwine like never before, a team of scientists at Carnegie Mellon University has taken a leap into the future by teaching robots to mimic the iconic celebrations of sports legends. Amongst the stars whose moves have been immortalised in metal and code are Cristiano Ronaldo, Kobe Bryant, and LeBron James.
The robots, trained meticulously using clips of the athletes' routines, have managed to capture the essence of these complex movements. Ronaldo’s famous “siu” celebration, a dynamic display of athleticism involving a jump, a 180-degree spin, and a triumphant arm slash, has been replicated with surprising accuracy. This is no small feat, considering the balance and agility required—traits that have historically been challenging for bipedal robots.
The study utilised the £14,000 Unitree G1 Humanoid robot, and the results have been promising. The new training method has halved the motion-tracking errors, a significant step forward in the quest for robots that can move with the fluidity and grace of their human counterparts. The robots, dressed in the jerseys of the sports stars they emulate, have not only managed to capture the physicality of these movements but also a semblance of their spirit.
This development is more than just a novelty; it represents a significant advancement in the field of robotics. The ability to transfer complex, agile movements from simulation to real-world application opens up a plethora of possibilities for humanoid robots. The scientists involved in the project have expressed optimism about the future, suggesting that this could pave the way for versatile robots capable of performing a wide range of tasks in real-world settings.
Meanwhile, the world of robotics continues to evolve at a rapid pace. Elon Musk’s Tesla has also entered the fray with its own humanoid robot, Optimus. Musk envisions a future where these robots become as commonplace as cars, performing tasks ranging from teaching to dog walking. While this vision is ambitious, the progress made by the team at Carnegie Mellon University suggests that such a future may not be as far off as it seems.
Comments