Embodied AI FRONTIER
/ Daily / NTU, PKU and Partners Unveil ω-0 World Action Model to Transform Humanoid Home Operations
Industry 📡 钛媒体

NTU, PKU and Partners Unveil ω-0 World Action Model to Transform Humanoid Home Operations

Home environments pose the ultimate test for humanoid robots. Unlike factories with structured and predictable pipelines, homes are dynamic and cluttered. A robot must navigate around furniture, adapt to moving targets, and maintain balance while performing tasks such as retrieving objects from the fridge or wiping a table. These seemingly mundane chores challenge the robot's perception, planning, and control capabilities.

To address these challenges, a research team from Nanyang Technological University, Peking University, HKUST (Guangzhou), and Beijing Academy of Artificial Intelligence introduced ω-0, a world action model based on latent-space prediction. This model takes language commands as input, simultaneously recognizes the environment and its own state, and directly outputs low-level action features that the control system can execute. This enables whole-body coordination for 'move-while-operate' tasks.

The team conducted extensive real-world evaluations on 11 highly difficult home operation scenarios, focusing on dynamic coordination, environmental adaptation, and autonomous decision-making. The tasks included cross-area object sorting, cleaning high and low positions, and picking up clothes from beds. The results were impressive: ω-0 Ego achieved a 79.1% success rate in first-person view, while ω-0 Omni reached 81.8% in full-scene mode. These figures significantly outperform established baselines such as π-0.5, EgoVLA, GR00T-N1.7, and ψ-0.

A key innovation of ω-0 is its rejection of the conventional video-to-action conversion, which is time-consuming and often inefficient. Instead, it adopts a unified query-based representation. In this architecture, future visual features provide high-level guidance about task progress and scene evolution, while action branches directly generate whole-body motor commands. This approach reduces latency and increases robustness, allowing the robot to adapt to unexpected changes during task execution.

To develop this model, the team designed a three-stage progressive training pipeline. The first stage involves dedicated pre-training for whole-body action VLM. The second stage is human-robot action latent fusion pre-training, which draws on advanced techniques from V-JEPA, integrating visual, linguistic, and proprioceptive data. By using video queries to predict future environmental features, the model embeds dynamic scene information into action representations. The third stage is fine-tuning with real-world data, where the model is exposed to a large corpus of home manipulation data, enabling it to adapt to diverse home environments.

Supporting this training effort, the researchers open-sourced ω-HOME, a large-scale dataset collected from real home settings. The dataset includes 40.3 hours of real environment data, 4,827 task trajectories, and 24 distinct tasks. Each trajectory contains synchronized language instructions, first-person and third-person camera views, proprioceptive states, and whole-body motion trajectories. This dataset significantly reduces the dependence on demonstration data for downstream tasks.

The 81.8% success rate is a remarkable achievement, but it is only a stepping stone. The journey toward a robot that can seamlessly integrate into daily life and operate autonomously over extended periods is still long. Nevertheless, ω-0 points the entire industry in a clear direction: treating the robot's body as an integrated whole rather than separate components. By learning to operate while moving and adjusting posture dynamically, robots can transition from laboratory demonstrations to practical home services, bringing us closer to the dream of embodied AI in everyday life.

✓ Verified Read Original → 2026-08-12
Flash Briefs Live
Today · Aug 18 Tue 18
15:15
Capital Agility 83
Agility CFO talks IPO prep and scaling with CFO Brew
11:47
Industry 东方财富 机器人概念 92
Xiaomi Q2 Revenue Hits 108.9B Yuan, Robot Debut and New Chip Announced
11:24
Capital The Robot Report 92
FORT Robotics to Take Safety Stack Public via SPAC Merger
11:13
Industry Booster Robotics 95
Booster Robotics Previews World Humanoid Robot Games
10:20
Industry AGIBOT 92
AGIBOT Hosts Jordan's King, Showcases Humanoid Robots & Embodied AI
09:51
Tech 华尔街见闻 快讯 92
Zhejiang Humanoid Robot Center Announces 'Co-evolution' Path for Real-World Deployment
08:55
Industry LimX Dynamics 92
LimX Dynamics to Showcase Embodied AI Solutions at WRC 2026
08:49
Industry The Robot Report 96
Unichem's Loomia Buy Accelerates Humanoid 'Skin' Market Entry
07:24
Industry 华尔街见闻 快讯 92
Shiyun Circuit: Humanoid Robot PCBs Now in Batch Supply
06:04
Industry Booster Robotics 95
Booster Robotics to Exhibit at IFA 2026 in Berlin
05:44
Industry 东方财富 机器人概念 96
Shangwei New Material Swings to Loss, Targets 4B Yuan Robot Revenue in 3 Years: Feasible?
05:37
Industry 华尔街见闻 快讯 96
Xiaomi MiMo-V2.5 Tops OpenRouter Weekly Calls; Robot Team Wins Dual Championships
04:37
Industry NEURA Robotics 92
NEURA Robotics and Schaeffler CEO Talk Scaling Humanoids to Millions
03:24
Capital 华尔街见闻 快讯 96
Perpetual Contracts Imply 300%+ Upside for Robot Maker Unitree's IPO
02:51
Capital 东方财富 机器人概念 95
Jiateng Robot Raises Strategic Funding to Strengthen Embodied Intelligence Chassis
01:57
Capital TechNode 92
Mech-Mind Passes Hong Kong Listing Hearing as Overseas Revenue Tops 50%
01:52
Policy 东方财富 机器人概念 92
Policy and Market Drive Robot Industry: Tax Support Boosts Beijing's Robotics Sector
01:41
Industry TechNode 92
Unitree Unveils 'Superman' Humanoid with 12.66 m/s Speed and 2-Meter Jump
Yesterday · Aug 17 Mon 33
23:22
Capital 东方财富 机器人概念 96
Unitree Tech to List on STAR Market Aug 19 Amid Robot Sector Rally
23:14
Industry AGIBOT 96
AGIBOT to Showcase Humanoid Robots at IFA 2026
23:14
Industry 东方财富 机器人概念 98
Stanley Robotics Equity Incentive Sparks Controversy as Executive Pay Soars
23:13
Industry Noetix Robotics 96
Noetix Robotics Shows E1 Robot Performing Cartwheels
22:48
Capital 钛媒体 98
Unitree Shifts PR Strategy to Superhuman Feats Ahead of IPO
22:22
Capital 东方财富 机器人概念 95
Jiateng Robotics Secures Tens of Millions in Strategic Funding
22:06
Capital 东方财富 机器人概念 96
Jiateng Robotics Secures Tens of Millions in Strategic Funding
17:15
Industry The Robot Report 92
Diligent Robotics rolls out Moxi 2.0 after five years of hospital deployments
15:34
Industry Agility 88
Agility Co-founder: Humanoid Robots Amplify Work, Not Replace It
15:32
Tech The Robot Report 88
Generalist leverages human demonstrations to accelerate robot learning
12:47
Industry 华尔街见闻 快讯 85
Hong Kong Firm Launches Humanoid Robot 'Xiaorui G3' for Heavy-Lifting Tasks
11:33
Industry IEEE Spectrum Robotics 88
Shipyard Welding: A Viable First Job for Humanoid Robots?
11:00
Industry Noetix Robotics 85
Noetix Robotics Shows Robot N2 Dancing to Music
09:49
Capital 东方财富 机器人概念 84
Shangwei Reports Revenue Growth but Stagnant Profits After Zhiyuan Takeover
09:47
Tech LimX Dynamics 83
LimX Dynamics TRON 2 Shows Autonomous Utility Tunnel Inspection
09:00
Tech Robohub 84
Intermittent Swimming Boosts Energy Efficiency of Fish-like Robots
09:00
Capital The Robot Report 88
Gravis Robotics Secures $200M Series A for Autonomous Construction
08:26
Industry 东方财富 机器人概念 88
Robot-Led High-Tech Manufacturing Grows 8.5 Points Faster Than Industrial Average
08:16
Capital 华尔街见闻 快讯 84
Unitree Robotics to List on STAR Market on August 19, 2026
06:43
Industry Galaxea Dynamics 85
Galaxea to Showcase Embodied AI at WRC 2026 in Beijing
06:14
Industry PaXiniTech 85
PaXini unveils PX6AX GEN4 tactile sensing units, to debut at WRC
05:53
Tech 东方财富 机器人概念 85
Guotai Junan Initiates Buy on Luoshi Robotics, Citing Embodied AI Sales Growth
04:56
Capital 华尔街见闻 快讯 84
Silin Zhiqu Plans Up to 1.8B Yuan Raise for Smart Manufacturing Base
04:56
Industry 华尔街见闻 快讯 88
Jiemei Tech to Invest Extra $428M in North China Base to Boost MLCC Release Film Capacity
04:06
Industry AGIBOT 88
AGIBOT to Host Partner Conference at LEAP 2026 in Riyadh
03:56
Industry TechNode 88
Robot Companies Transform into AI Firms as AgiBot Redefines Embodied AI Competition
03:55
Industry 钛媒体 84
Taikang Insurance Heir Leads $274M Bet on Unitree in Hard-Tech Push
03:48
Industry Galaxea Dynamics 83
Galaxea Dynamics to Host Global Partner Night at WRC 2026
03:29
Industry 东方财富 机器人概念 88
Unitree Launches 'Superman' Robot, Output Tops 18,000 Units, ETF Gains 2%
03:05
Industry 东方财富 机器人概念 88
STAR 200 ETF Soars 4.2% as Tech Rally Expands to Semiconductors, CPO, Robots
02:38
Industry Unitree 91
Unitree Unveils 'Superman' Robot with 2m Jump, 12.66m/s Speed
02:03
Industry 华尔街见闻 快讯 85
Unitree Unveils Humanoid Robot That Breaks Jump and Speed Records
01:39
Industry TechNode 88
Volkswagen China to Roll Out Advanced Driver Assistance Across Three JVs
Embodied AI FRONTIER