RoboSignal Daily · 2026-10-22 · Thursday
ROBOSIGNAL
NVIDIA Releases Cosmos3-H-Embodied-Nano for Surgical Robotics
NVIDIA has released Cosmos3-H-Embodied-Nano, a multimodal model trained on surgical robotics data for forward dynamics, inverse dynamics, and policy prediction. It is based on the Cosmos3-Nano model and is intended for use in surgical world modeling and imitation learning.
Robot Models
NVIDIA Releases Cosmos3-H-Embodied-Nano for Surgical Robotics
NVIDIA has released Cosmos3-H-Embodied-Nano, a multimodal model trained on surgical robotics data for forward dynamics, inverse dynamics, and policy prediction. It is based on the Cosmos3-Nano model and is intended for use in surgical world modeling and imitation learning.
Research
LoopWAM: Closed-Loop Generalist Policy Improvement from Deployment Rollouts
LoopWAM, a world action model developed by Tsinghua University, Yuanli Lingji, and Jihuan Shidai, achieves top performance on RoboChallenge Table30 V2 and other benchmarks. It leverages real-world deployment data to iteratively improve robot policies through quality-guided learning, incorporating success, suboptimal, and failure trajectories to enhance world modeling and action generation. The model uses a video-action dual-branch architecture and quality steering to guide policy improvements, achieving significant performance gains across multiple tasks.
NeuroVLA: The Global First Class-Brain Architecture VLA Embodied Large Model
Zhi Fang Ping has developed NeuroVLA, the first class-brain architecture VLA embodied large model, featuring a 'cerebral cortex-cerebellum-spinal cord' three-layer system. It demonstrates safety reflexes in 20 milliseconds, 80.2% reduced jerk, and 0.4W power consumption. The company has achieved commercial deployment across 7+ industries, with 2000 annual production capacity and a 1000-unit order from Huiké. The model's progress is supported by data and industry recognition, though real-world deployment and long-term viability remain to be validated.
arXiv Robotics — research abstractsFrom Instrument-Mounted Demonstrations to In-Vivo Execution: Learning Bimanual Laparoscopic Appendectomy Without Robot-Collected Demonstrations
This research presents an end-to-end pipeline for training a surgical robot policy using motion data captured from standard laparoscopic instruments. The system uses inertial, time-of-flight, and Hall sensors to track instrument state without external cameras. A diffusion policy is trained on both ex-vivo and in-vivo demonstrations, achieving appendectomy in three of four live rabbits. The robot serves as a timing reference and executor, with no demonstration collection.
RoboParty Releases RP1, Huang Yi Aims to Demystify Embodied Intelligence
RoboParty has officially released RP1, a humanoid robot designed for developers, emphasizing full-stack open-source. The robot includes mechanical designs, firmware, control systems, and training tools, aiming to create a platform for developers to build upon. The release has sparked significant interest on platforms like X, with developers expressing enthusiasm for the open-source approach.
arXiv Robotics — research abstractsRoboEdit: Turning Human Manipulation Videos into Scalable Robot Experience
This paper presents RoboEdit, a system that transforms human manipulation videos into physically plausible robot videos with aligned 3D hand states. The method includes a pipeline for cross-embodiment adaptation and a large-scale dataset of 174K video pairs. Experiments show state-of-the-art editing quality and support for real-world robot control policies.
Deep Planning Intelligent's Four-Legged Robot 'Fumo' Wins Grid Power Project
Deep Planning Intelligent has secured a contract with State Grid Power for its four-legged robot 'Fumo' in a complex environment project. The robot is designed for inspection and maintenance in challenging power infrastructure settings, featuring advanced environmental sensing, navigation, and task execution capabilities. The project highlights the growing demand for specialized robotics in high-value industries.
GPT-6-Astra's Embodied Intelligence Under Scrutiny: High Scores in Simulations, Struggles in Real-World Execution
A research team from SenseNova and collaborators conducted a comprehensive analysis of GPT-6-Astra's embodied intelligence capabilities, achieving high scores in simulation benchmarks but facing significant challenges in real-world robot execution. The study highlights the gap between simulated and real-world performance, emphasizing the need for improved physical interaction, task planning, and execution efficiency.
Robotics in Logistics: The Rise of Non-Humanoid Sorting Robots
The article explores the deployment of non-humanoid robots in logistics, focusing on parcel sorting. It discusses the technical and economic advantages of these systems, including their ability to handle high-frequency, non-structured tasks, and the shift from demonstration to real-world deployment. Key players like StarDance, SelfVariable, and Jiajia Vision are highlighted, along with their approaches to automation in logistics scenarios.
ADI Creates 'Neural Endings' for Humanoid Robots
ADI provides chip solutions for force sensing, tactile feedback, and power protection, offering foundational support for perception and execution layers in humanoid robots. These technologies, originally developed for automotive and drone applications, have been optimized for use in robot joints, fingers, and power systems, addressing space constraints and high-voltage management issues. ADI does not engage in full robot manufacturing but focuses on solutions for the 'neural endings' of robots, providing the industry with mature and reliable hardware support.
JiZhiJia Invests 50 Million RMB in ShenZhi, Debuts Two Robots at the 2026 Industrial Fair
JiZhiJia invested 50 million RMB in ShenZhi Robotics, acquiring a 20% stake. ShenZhi unveiled its self-developed embodied base model SE-1 and two robots at the 2026 Industrial Fair: Sculptor for multi-task heavy-duty logistics operations, and Edge for industrial precision manipulation.
Deployments
Robotics in Logistics: Deployment of Non-Humanoid Sorting Robots
The article explores the deployment of logistics robots in warehouse sorting tasks, focusing on non-humanoid designs that are optimized for high-frequency, data-rich environments. It discusses the challenges of human labor in sorting, the technical and economic rationale for robotic solutions, and the evolving role of embodied AI in logistics automation.
Briefs
- Shenmou's Four-Legged Manipulation Robot 'Fumo' Wins Grid Power TenderRobotics — Chinese web discovery
- Chinese Smart Lawn Mowing Robots Compete in Global MarketChinese robotics — Company news discovery
- Real-World Training Platforms: The Next Infrastructure for Embodied IntelligenceRobotics — Chinese web discovery
- Sinovation Tests GPT-6-Astra and Opens Sources SenseNova-RoboRSI: What's Next for Embodied Intelligence?Robotics — Chinese web discovery
- Learning Where it Matters: Autonomous Data Collection and Offline Reinforcement Learning for Local Contact-Rich ManipulationarXiv Robotics — research abstracts
- 00-Post Gen Entrepreneurs' Embodied Intelligence Exploration: From Bipedal Robots to Home CleaningChinese robotics — Research news discovery
- Robotics Deployment Discovery: From Performance Rentals to Content-Driven EntertainmentChinese robotics — Deployment discovery
- Blue Horizon's Spatiscan S1: A 3D Camera for Robust Perception in Non-Structured Outdoor EnvironmentsRoboSpeak — WeChat
- Human-Robot Ping-Pong at Canton Fair: A Testbed for Embodied IntelligenceChinese robotics — Robot learning research
- Zhiyuan's Full-Sized Humanoid Robot Retail Pod Launched in Wuhan JiangtanChinese robotics — Hardware and sensing