In April 2026, nearly 100 humanoid robot teams and over 300 robots, including Shandong Youbaote Intelligent Robot Co., Ltd., gathered at the Beijing Yizhuang Half Marathon, also serving as the humanoid robot half-marathon racecourse.

In this event, CHINGMU provided invaluable technical support to Youbaote, comprehensively capturing the running postures of amateur sports enthusiasts. Through data optimization, repeated testing, and nearly 100 hours of simulation training in virtual environments, the humanoid robot ultimately achieved a highly human-like, coordinated, and smooth running gait. CHINGMU's fully self-developed high-precision motion capture system, with sub-millimeter accuracy of ±0.02mm, completely captures the body movement data of professional marathon runners during their runs . It also supports integration with 3D force plates, instrumented treadmills, and other equipment to simulate various walking, running, and complex environments. By performing spatial tracking on the robot's key structures and obtaining real-time 6DoF pose data, it analyzes dozens of quantitative indicators including postural stability, trajectory accuracy, center-of-mass deviation error, and turning response speed, allowing for early validation and optimization of robot performance under extreme road conditions.

Meanwhile, autonomous navigation technology is gradually becoming an industry trend, placing higher demands on motion capture technology for pre-race training. As a global leader in motion capture, CHINGMU supports the entire process from motion data acquisition and dataset construction to robot training, control algorithm validation, teleoperation real-time driving, and full-scenario performance evaluation—building a motion capture solution that covers the entire R&D lifecycle of humanoid robots, providing foundational data technology support for the development of the embodied AI industry .