While China has dominated the headlines for humanoid robots, South Korean companies have introduced humanoids that learn complex movements through open source AI frameworks.
In a recent demonstration, ROBOTIS’ AI Sapiens learned the famous CORTIS REDRED Challenge motions using only smartphone video, eliminating the need for a professional motion capture system.
This process combined video-based motion capture, motion retargeting, simulation-based reinforcement learning, and Sim2Real transfer.
According to the company, the demonstration highlights how the open-source tool simplifies humanoid robot training, allowing users to more easily generate, learn, and perform full-body movements.
Humanoid learns movement
ROBOTIS has demonstrated the capabilities of the AI Sapiens humanoid robot, an open source platform for physical AI, equipped with DYNAMIXEL-Q actuators. This project aims to make behavioral learning for humanoid robots more accessible using widely available hardware and open-source software tools.
In the demo, AI Sapiens learns and performs complex full-body movements known as the CORTIS REDRED Challenge. Instead of relying on expensive professional motion capture systems, the robot learns its movements from video recorded using a standard smartphone camera. This significantly reduces the cost and complexity of collecting training data for humanoid robots.
The motion learning process begins with video-based motion capture. Human movements recorded on a smartphone are converted into digital motion data that can be processed by software. The captured motion passes through a motion retargeting stage, which adjusts the human movement to match the humanoid robot’s physical structure and joint limitations.
After retargeting, the robot is trained in a simulated environment using reinforcement learning. At this stage, the AI can practice movements repeatedly in the virtual world to improve balance, coordination, and precision of movement without risking damage to physical hardware. Simulation training also allows for rapid testing and optimization of motions before deploying them to a real robot.
Once training is complete, the learned behaviors are transferred from the simulation to the physical AI Sapiens robot through the Sim2Real pipeline. This process helps ensure that motions developed in a virtual environment can successfully execute in the real world, despite the differences between simulation and physical hardware.
Accessible AI robotics
ROBOTIS plans to release its motion generation and learning pipeline as open source software, giving researchers, developers, educators, and hobbyists access to the tools used in the demonstration. The goal is to lower the barrier to humanoid robot development and enable the broader community to experiment with behavioral learning and physical AI systems.
According to ROBOTIS, AI Sapiens is a fully open-source humanoid robot platform designed for physical AI research and development. At 1.3 meters tall and weighing 34 kilograms, the robot has 23 degrees of freedom throughout its body, allowing it to perform a wide range of human-like movements.
The platform is equipped with 23 next-generation DYNAMIXEL-Q semi-direct drive (QDD) actuators, including 14 QM-060 units and 9 QM-080 units. These actuators combine low gear reduction ratios, high-torque motors, and integrated control electronics to provide high backdriveability, low impedance, and precise torque control, making them suitable for dynamic and docile humanoid movements.
AI Sapiens is powered by an NVIDIA Jetson Orin NX 16GB computer, delivering up to 100 TOPS of AI performance for advanced robotic tasks. It supports Wi-Fi 5, Bluetooth 5.0, dual Ethernet ports, USB connectivity, and a 24V/12V power output for connecting additional hardware.
The robot is powered by a 46.8V, 9000mAh battery. It is supported by a complete open source ecosystem including hardware bill of materials, CAD files, source code, simulation assets, and development tutorials, allowing researchers and developers to customize and extend the platform.
Jijo is an automotive and business journalist based in India. He holds a BA (Hons) in History from St. Stephen’s College, University of Delhi and a PG degree in Journalism from the Indian Institute of Mass Communication, Delhi.He has worked in news agencies, national newspapers and motoring magazines. In my free time, I like to go off-roading, participate in political discussions, travel, and teach languages.
