Hong Kong Physical AI Stocks Rise as Apple Unveils HALO Humanoid Control System
港股物理AI概念走强，苹果披露人形机器人自主控制系统HALO丨港股异动
On November 9, Hong Kong stocks linked to physical AI gained ground, with Kailuishi Technology up 11.58%, Hesai-W rising 6.58%, Xunce increasing 3.64%, and Qunhe Technology climbing 2.91%. On November 4, Apple's research team submitted a preprint on arXiv titled 'Perceptive Humanoid Motion from Simulation', disclosing the HALO autonomous control system for humanoids and releasing open-source code for data generation, model training, and deployment across action generation, tracking, task orchestration, simulation, and real robot operation. The study reports an average success rate of 91% across 360 single-task tests in six unscanned environments and 70% for three combined tasks in one environment.
Source: Chinese robotics — Research news discovery · Read original article ↗
English translation unavailable. Choose the original article to read it.
What the source reports
Publisher-reported claims, with original evidence. These results have not been independently verified by RoboSignal.
- Control
- Reported autonomousOpen source S4
Reported numbers
Reported success rate
91%
Unique elapsed hours
View original evidence
据论文披露，研究团队在6个未预先扫描或建图的真实环境中开展360次单项任务测试，平均成功率为91%；另在一个真实环境中，对3种组合任务分别开展10次测试，平均成功率为70%。
Open source S5Reported success rate
70%
Unique elapsed hours
View original evidence
据论文披露，研究团队在6个未预先扫描或建图的真实环境中开展360次单项任务测试，平均成功率为91%；另在一个真实环境中，对3种组合任务分别开展10次测试，平均成功率为70%。
Open source S5test_count
360
Reported trials
View original evidence
据论文披露，研究团队在6个未预先扫描或建图的真实环境中开展360次单项任务测试，平均成功率为91%；另在一个真实环境中，对3种组合任务分别开展10次测试，平均成功率为70%。
Open source S5test_count
10
Reported trials
View original evidence
据论文披露，研究团队在6个未预先扫描或建图的真实环境中开展360次单项任务测试，平均成功率为91%；另在一个真实环境中，对3种组合任务分别开展10次测试，平均成功率为70%。
Open source S5robot_count
21,200 robots
View original evidence
数据是当前最核心瓶颈和机会，据中国信通院测算，机器人迎来"ChatGPT时刻"约需2.12万台机器人连续工作一年、产出千万小时级有效数据，而当前全球高质量真实数据仅数十万至百万小时级，真机数据缺口显著。
Open source S9Reported success rate
95%
Reported trials
View original evidence
华源证券表示，Astra初步展现通用大模型潜力，第三方测试显示其在Robocurve简单抓放任务中成功率达到95%；Astra与专用VLA模型π0.5协同后，复杂双臂操作成功率由26%提升至48%。
Open source S10Reported success rate
48%
Reported trials
View original evidence
华源证券表示，Astra初步展现通用大模型潜力，第三方测试显示其在Robocurve简单抓放任务中成功率达到95%；Astra与专用VLA模型π0.5协同后，复杂双臂操作成功率由26%提升至48%。
Open source S10
- paper: not_reported
- code: not_reported
- weights: not_reported
- dataset: not_reported
What remains unknown
Not established in the collected evidence: Environment, Data origin.
Reported performance applies to the described task. It does not establish general autonomy or deployment readiness.
Source excerpts and review record
Automatically extracted; no manual editorial approval recorded.
近日，据澎湃新闻报道，当地时间11月4日， 苹果 研究团队在预印本平台arXiv提交论文《Perceptive Humanoid Motion from Simulation》，披露了人形机器人自主控制系统 HALO ，并公开了数据生成、模型训练和部署代码，涵盖动作生成、动作跟踪、任务编排以及仿真和真实机器人运行等环节。
Open source S4
据论文披露，研究团队在6个未预先扫描或建图的真实环境中开展360次单项任务测试，平均成功率为91%；另在一个真实环境中，对3种组合任务分别开展10次测试，平均成功率为70%。
Open source S5
数据是当前最核心瓶颈和机会，据中国信通院测算，机器人迎来"ChatGPT时刻"约需2.12万台机器人连续工作一年、产出千万小时级有效数据，而当前全球高质量真实数据仅数十万至百万小时级，真机数据缺口显著。
Open source S9
华源证券表示，Astra初步展现通用大模型潜力，第三方测试显示其在Robocurve简单抓放任务中成功率达到95%；Astra与专用VLA模型π0.5协同后，复杂双臂操作成功率由26%提升至48%。
Open source S10
Source：Chinese robotics — Research news discovery · news.qq.com