Video: Humanoid robot learns to autonomously spar and fight without human control

Unitree Robotics has demonstrated a humanoid robot fighting autonomously, marking a step toward AI systems that can control robots without human operators. The company’s G1 humanoid used a new AI model, named UnifoLM-X2-1.0, to spar with a human partner in real time. The demonstration highlights growing efforts to develop robots that can respond to rapidly changing physical environments while planning and executing movements on their own. Unitree says the system represents a shift from earlier humanoid combat demonstrations that relied on remote control or other forms of human intervention. G1 gets autonomous A new video shows Unitree’s G1 humanoid sparring autonomously against a human partner using the UnifoLM-X2-1.0. The system is designed to eliminate the human teleoperator typically required to control humanoids during combat demonstrations. Instead of converting commands from a VR headset, gamepad, or other controller into movements, UnifoLM-X2-1.0 uses a predictive perception-to-action system. It continuously processes the environment, anticipates how the scene and opponent are likely to change, and uses those predictions to plan the robot’s next movements. In demonstration footage, the G1 responds to an approaching human opponent by adjusting its stance and footwork, avoiding incoming strikes, and executing punches and kicks. Visualizations in the footage highlight the system’s predictive modeling, showing how future states of the interaction are incorporated into motion planning before the robot physically moves. The technology targets a major challenge in deploying AI-powered humanoids: reducing the delay between perception, decision-making, and physical action. Unitree says UnifoLM-X2-1.0 is designed to address bottlenecks in world-action models by enabling faster motion planning, tactical decisions, and closed-loop execution during highly dynamic interactions. “UnifoLM-X2-1.0 breaks through world-action foundation models’ bottlenecks in instant planning, decision-making, and dynamic interactive execution, achieves high dynamics, strong interaction, real-time prediction and planning of the future, achieves fully autonomous combat for humanoid robots,” read the video description on YouTube. Gets predictive AI Unitree’s latest demonstration highlights how its humanoid robots are moving from remote control toward more advanced autonomous systems. The technology behind the G1’s autonomous sparring is part of Unitree’s growing family of foundation models, which has evolved from models designed for factory manipulation to newer multimodal systems. UnifoLM-X2-1.0 uses a world-model approach to help the robot understand its surroundings and predict what could happen next. Experts suggest that this is important in fast-moving situations such as combat, where simply reacting to what a robot sees may be too slow. The system can predict an opponent’s movement while also considering the G1’s own balance and position. It can then use these predictions to select movements and adjust its actions in real time. However, footage from the demonstration suggests that the system may not be running entirely on the robot itself. Console logs showing commands such as OBS/replan and env.step indicate that sensor information may be sent to a separate computer running the AI model, which then sends movement instructions back to the G1. This means reducing the system’s reliance on external computing remains an important challenge.Unitree sees autonomous combat as a test for physical AI. If robots can predict movements, maintain balance, and respond to unexpected contact, similar technology could eventually help humanoids work more safely in complex industrial environments.Get the latest in engineering, tech, space & science - delivered daily to your inbox.Jijo is an automotive and business journalist based in India. Armed with a BA in History (Honors) from St. Stephen's College, Delhi University, and a PG diploma in Journalism from the Indian Institute of Mass Communication, Delhi, he has worked for news agencies, national newspapers, and automotive magazines. In his spare time, he likes to go off-roading, engage in political discourse, travel, and teach languages.