The 2026 Beijing Humanoid Robot Half Marathon has officially concluded. Compared with the 2025 event, this year’s competition demonstrated significant advancements in scale, technical complexity, and ecosystem participation—highlighting the industry’s accelerating transition toward commercialization.
According to IDC, global shipments of humanoid robots are expected to exceed 510,000 units by 2030, representing a compound annual growth rate (CAGR) of nearly 95%. As foundational technologies mature, application value deepens, ecosystems expand, and business models evolve, competition will increasingly center on real-world deployment capabilities and value delivery.
This year’s event attracted over 100 teams from enterprises, universities, and research institutions. There were notable breakthroughs in both core technologies and product performance. Leading players—including Honor—delivered standout results, with some robots surpassing human-level running speeds, reflecting significant improvements in locomotion capabilities.
Honor’s championship win underscores a broader industry trend: growing strategic commitment to humanoid robotics. As embodied intelligence continues to evolve, computing power, algorithms, and foundational models remain critical. Meanwhile, the entry of consumer device vendors is expected to create differentiation in edge computing and visual models. Combined with their large user bases and ecosystem strengths, they are well-positioned to accelerate both technological iteration and real-world adoption.
The competition has moved beyond testing basic mobility. Notably, 38% of participating teams adopted fully autonomous navigation, and the winning robot leveraged this capability under weighted scoring rules.
This shift reflects a broader transition: humanoid robots are now evaluated by their ability to operate reliably in complex, dynamic environments. Success requires seamless integration across perception, decision-making, and executions supported by high system stability and engineering readiness for scalable deployment.
Multimodal Perception
Robots combined data from multiple sensors—including satellite positioning, LiDAR, vision systems, and IMUs—along with real-time mapping technologies. This enabled stable operation across slopes, sharp turns, uneven terrain, and dynamic obstacles, significantly enhancing adaptability to real-world environments.
Reinforcement Learning and Motion Control
Extensive reinforcement learning in simulation environments, combined with high-quality human motion data and continuous adaptation and tuning in real-world environments, has driven major improvements in motion control algorithms. Robots now demonstrate enhanced balance, posture optimization, and human-like movement, enabling long-distance autonomous navigation, obstacle avoidance, and path optimization.
Hot-Swappable Batteries and Liquid Cooling
Energy management emerged as a critical enabler. Hot-swappable battery systems allow efficient recharging without downtime, while advancements in lightweight design, energy optimization, intelligent power distribution, and liquid cooling significantly extend operational endurance.
Hardware-Software Co-Optimization
Breakthroughs in hardware-software integration and engineering capabilities emerged as a key highlight and a critical focus for future development. Facing real-world physical environments and task requirements, robots must continuously learn and adapt to new scenarios and tasks, enabling end-to-end coordination across perception, decision-making, and execution. At the same time, deeper integration and precise alignment between AI models and diverse hardware configurations are required to establish real-time interaction loops. This ensures responsiveness, control precision, and system stability in complex tasks, accelerating progress toward highly reliable, maintainable, and scalable real-world deployment.
Humanoid robotics is rapidly evolving toward a closed-loop embodied intelligence system encompassing perception, learning, decision-making, and execution. Engineering capability is emerging as the key determinant of commercial viability.
In 2025, the global humanoid robot market experienced a breakout year, led by Chinese vendors, with shipments exceeding 18,000 units. More than 85% of deployments were concentrated in performances, education, data collection and guided tour service scenarios—primarily focused on demonstration, interaction, and technology validation. Early pilots have also emerged in manufacturing and logistics.
Looking ahead, IDC Humanoid Robotics Research forecasts that by 2030:
Chinese vendors are expected to account for 95% of global shipments in 2025, establishing a dominant position in manufacturing and scalability. Several leading companies are projected to achieve annual production capacity in the tens of thousands by 2026, further strengthening supply capabilities.
Key innovation areas include structural optimization, joint and energy system upgrades, mass production capabilities, and motion control improvements. Additionally, dexterous hands—critical for fine manipulation—are poised for rapid development.
Fastest-growing categories:
Industrial Adoption Accelerates
Collaboration with industrial leaders is validating performance metrics such as cycle time, task success rates, and operational stability. Deployment in manufacturing environments is expected to scale rapidly, with shipment growth exceeding 200% in 2026.
IDC research indicates that over the next three years more than 80% of users plan to deploy robots in tasks such as palletizing, handling, picking, and machine tending.
Service Applications Deepen
Humanoid robots are expanding into personalized services, enhancing customer experience and engagement in areas such as retail guidance and food service.
Data Scale Expansion
The integration of simulation data, internet video data, and real-world operational datasets is driving rapid growth in training data. China has already accumulated tens of thousands of hours and nearly petabyte-scale datasets, andis leading the development of the world’s first international standard for humanoid robot datasets.
Model Evolution
Advancements in motion models, combined with deeper integration of world models and vision-language-action (VLA) models, are improving generalization and intelligence. Leading global humanoid robotics and AI model companies are continuously accelerating the iteration and upgrading of foundational models.
Scenario Co-Creation
Oriented toward real-world industrial application scenarios, ecosystem players are jointly advancing application development and solution building, bridging the critical gap between technical validation and large-scale deployment, and accelerating the industrialization of humanoid robotics technologies.
Robot-as-a-Service (RaaS) models—including leasing and subscription—are lowering adoption barriers and accelerating market penetration.
IDC research shows that user acceptance of RaaS has doubled year-over-year. As service systems, pricing models, and maintenance capabilities mature, adoption is expected to further accelerate.
IDC believes the 2026 Beijing Humanoid Robot Half-Marathon has become a key benchmark for assessing both technological maturity and industry progress. The event not only validated core capabilities but also accelerated ecosystem development and commercialization.
Over the next 2–3 years, the humanoid robotics industry will enter a pivotal phase:
Humanoid robotics is transitioning from technically feasible to commercially viable. Leading companies are already securing strategic positions in high-value scenarios, while the window for late entrants is rapidly narrowing.
This analysis is based on IDC research, including:
For organizations evaluating entry strategies, identifying priority use cases, or assessing vendor capabilities, IDC offers comprehensive research and advisory services to support decision-making and accelerate time-to-market. For more information and related research, please contact trago@idc.com.
Lily Li is a research manager for emerging technologies in IDC China. She is responsible for conducting research and analysis for Internet of Things (IoT) in the same country. She is also involved in global and regional consulting as well as business development in related markets. Prior to joining IDC, Lily has had in-depth working experiences in the urban digital transformation (DX) field and a wide range exposure to Smart City developments. She has a deep understanding of the status quo and is knowledgeable about the market's future trends. Lily holds a master's degree from the Graduate University of Chinese Academy of Sciences (GUCAS).
此内容由惯性聚合(RSS阅读器)自动聚合整理，仅供阅读参考。 原文来自 — 版权归原作者所有。