<ul><li>Researchers in China and Hong Kong have developed an AI framework that teaches humanoid robots to stand up quickly from an idle position.</li><li>The framework, called Humanoid Standing-up Control (HoST), uses reinforcement learning to allow the robot to learn through trial and error.</li><li>The HoST framework was first trained in simulations and then deployed in a Unitree G1 Humanoid Robot for experimental testing.</li><li>The successful results pave the way for integrating standing-up control into existing humanoid systems, expanding their real-world applicability.</li></ul>

Watch this humanlike robot 'rise from the dead' with creepy speed and stability

Discover more