RoboSignal Monthly · 2026-10
ROBOSIGNAL
Zhiyuan Releases Embodied Model GO-3 Preview, Tops Three Industry Rankings
本月 30 developments，最受关注的是：Zhiyuan Releases Embodied Model GO-3 Preview, Tops Three Industry Rankings；AGIBOT Open-Sources AGIB T WORLD Release 4: 'Physical Reasoning' Dataset Enabling Robots to Reason Before Acting；Scaling Online Exploration for Robot Agents。
Research
Zhiyuan Releases Embodied Model GO-3 Preview, Tops Three Industry Rankings
On October 29, 2026, Zhiyuan released the GO-3 Preview, an embodied model based on the native embodied architecture WRAM, achieving unified task reasoning, future prediction, and action generation. GO-3 Preview ranked first in three major rankings: RoboDojo, RoboTwin, and RoboColiseum, and plans to fully open-source by the end of 2026. The model continuously iterates through pre-training, post-training, and deployment, enabling end-to-end scalable learning and enhancing robots' adaptability and execution capabilities in complex tasks.
AGIBOT Open-Sources AGIB T WORLD Release 4: 'Physical Reasoning' Dataset Enabling Robots to Reason Before Acting
AGIBOT has released the fourth version of its AGIBOT WORLD dataset, focusing on 'Physical Reasoning' to enable robots to make decisions before acting. The dataset covers over 10 categories of logical reasoning scenarios, supporting both high-level decision-making models and low-level action policies. It includes annotations linking task goals to execution processes, aiding in the training of embodied agents for real-world tasks.
arXiv Robotics — research abstracts10-19Scaling Online Exploration for Robot Agents
This research presents RAX, an open-source framework for robot agents that enables continuous self-improvement through online exploration. The framework allows agents to acquire experience, synthesize executable programs, and perform complex tasks without human demonstrations. RAX demonstrates significant performance improvements in both simulation and real-world settings, with results showing a +37.4% average improvement over top baseline models.
Isaac Lab — ReleasesFirst-party10-20Isaac Lab 3.0 Released: Multi-Backend Architecture for Robot Learning
Isaac Lab 3.0 is now generally available, offering a unified framework for robot learning with support for multiple physics, rendering, and visualization backends. Key features include Newton physics, OV libraries, and Isaac Sim integration, along with improved simulation tools and reinforcement learning compatibility.
Science Robotics Latest Cover: 100 km/h. AI-Controlled Drone Beats Human World Champions
The MonoRace system, developed by the MAVLab team from Delft University of Technology in the Netherlands, defeated three human FPV racing world champions in the Abu Dhabi Autonomous Drone Racing Championship, achieving the fastest single-lap time of 16.56 seconds. The system, which uses only a monocular camera and IMU, achieved state estimation and control during high-speed flight through reinforcement learning and simulation training, demonstrating robustness under extreme conditions.
arXiv Robotics — research abstracts10-14GreenWAM: A World Foundation Model for Robotics
This paper introduces GreenWAM, a world foundation model for robotics that serves both world modeling and policy generation roles. It achieves top-tier performance on WorldArena 1.0 and ranks among the top entries in RoboTwin 2.0. The model is adapted for real-world humanoid robots and distilled into a single-step policy for real-time control, with controlled experiments highlighting the benefits of robot specialization and maintaining world-model modes during policy adaptation.
Tech Xplore — Robotics10-17Team develops embodied robotic arm 'EmArm' with human-like touch sensing
A research team at HKUST has developed EmArm, an embodied robotic arm capable of whole-arm tactile sensing and adaptive manipulation. By integrating a soft optical tactile skin with a rigid skeleton and real-time decision-making algorithms, EmArm achieves high tactile fidelity and can distinguish between different touch interactions, such as tickling and patting. The system also enables emotional feedback expressions and safe autonomous operation in complex environments.
Zero-Shot Real-World Deployment. RAI Institute Creates 5-Motor Bicycle Robot: 8 m/s, Half-Second Aerial Flips, 1-Meter Jump Platform
The RAI Institute's UMV robot, a bicycle-like machine with 5 motors, achieves 8 m/s speed, 1-meter jumps, and half-second aerial flips without specialized balancing hardware. All movement strategies are trained in simulation and directly deployed to the physical robot, demonstrating zero-shot transfer. The robot uses reinforcement learning to manage balance and dynamic maneuvers, with no springs, control moment gyroscopes, or dedicated stability mechanisms.
arXiv Robotics — research abstracts10-30Kinematic-Tactile Aligned Dexterous Retargeting with Contact Wrench Space Alignment
This research presents a three-stage framework for dexterous retargeting that aligns human and robot hand kinematics and tactile information through contact wrench space. The method achieves 96.7% replay success in real robot experiments, demonstrating effective transfer of physical interaction data for manipulation tasks.
arXiv Robotics — research abstracts10-27RynnVLA-Lat, Scaling Robotic Manipulation Foundation Models with 3D-Grounded Latent Actions
This research presents RynnVLA-Latent, a framework that enables scalable pretraining of robotic manipulation policies by learning 3D-grounded latent actions from heterogeneous interaction videos. The method avoids reliance on native robot controls and leverages visual transitions, future-feature prediction, and camera-motion decoupling to capture control-relevant motion. The results show strong performance on multiple benchmarks, including LIBERO, LIBERO-Plus, VLABench, and RoboTwin 2.0, with success rates exceeding those of competing methods in several real-world tasks.
arXiv Robotics — research abstracts10-12ForeAct3D: Policy-Grounded Future World Modeling for VLA Policies
This paper presents ForeAct3, a framework for policy-grounded future world modeling within VLA policies. It uses learnable geometric queries to decode depth, semantic segmentation, and camera pose from policy representations, enabling accurate prediction of future 3D scene states. The method improves manipulation performance through semantic 3D supervision, physical consistency, and action conditioning, achieving 98.3% average success on LIBERO and 37.8% success in real-world tasks.
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.
HandsOn: A Benchmark for Dexterous Manipulation with Cross-Embodiment and Task Generalization
A research team from Tsinghua University, in collaboration with other institutions, has developed the HandsOn benchmark, a comprehensive evaluation framework for dexterous robotic hands. The benchmark includes 66 tasks and 18,000 human-teleoperation demonstrations, focusing on tool use, fine contact, bimanual coordination, and task transfer. It evaluates the performance of various models, including diffusion policies and visual-language-action models, and highlights the challenges in achieving stable execution and generalization across different tasks and objects.
Clearpath Robotics — BlogFirst-party10-14IROS Research Round-Up: Husky and Jackal in Action
At IROS 2026, Clearpath Robotics showcased six research papers that leverage their Husky and Jackal platforms to address real-world robotics challenges. These include heterogeneous robot collaboration using generative AI, disease detection in orchards, robust LiDAR-inertial-visual odometry in coal mines, safety-constrained reinforcement learning for navigation, 4D-aware urban navigation from internet videos, and a novel calibration framework for event cameras and odometry.
Physical AI and Embodied Intelligence: Industry Signals Point Toward a Critical Inflection Point
The article discusses the convergence of physical AI and embodied intelligence, with key industry events and product launches signaling a shift toward standardized robot configurations and model-driven production. The release of H2-X and GPT-6Astra demonstrates progress in direct model control and physical interaction, while policy support and market data suggest the industry is approaching a critical mass.
Hong Kong University of Science and Technology Team Develops Helical Scaffold for Stable Bronchoscopic Imaging and Intervention
A team from the Hong Kong University of Science and Technology, led by Professors Ren Hongliang and Yuan Wu, has developed a helical brace that stabilizes robotic instruments during bronchoscopic procedures in breathing lungs. The device reduces tube displacement by 96% while maintaining airway patency. It was tested in simulated and in vivo pig models, showing effectiveness in bronchoscopic interventions and imaging. The study, published in Science Robotics, highlights the potential of the helical brace for clinical applications, though further research is needed before human use.
ETH Zurich — Robotics research newsFirst-party10-22One Small Leap for a Muscle, One Giant Step for Robotics
ETH Zurich researchers have created biohybrid robots with living muscle tissue and soft artificial scaffolds, enabling visible and powerful movements such as jumping, swimming, walking, and grasping. The integration of muscle and scaffold design allows for a range of motion up to eleven times greater than earlier systems, though long-term deployment remains a challenge.
arXiv Robotics — research abstracts10-26See What I Need, Dream What I Want: A Navigation Foundation Model on a Self-Constructed Representation Space
This paper presents RepresentNav, a navigation foundation model that constructs a self-constructed representation space using a compact three-layer transformer. It learns to see what it needs for navigation and predicts future scene representations through a joint-embedding predictive architecture, achieving state-of-the-art results on multiple benchmarks and physical robot navigation tasks.
arXiv Robotics — research abstracts10-09SDPAD: A Fully Spike-Driven Pipeline for End-to-End Autonomous Driving
This paper introduces SDPAD, a fully spike-driven end-to-end planning pipeline for autonomous driving. It converts pre-trained ANN perception stacks into integer-spike form, lifts multi-view images into BEV space using Spike-3D-Lift, and plans through Spike-QFormer, a spiking query transformer. SDPAD achieves high accuracy and low energy consumption, outperforming previous SNN planners and matching mainstream ANN planners in energy efficiency.
Robotics & Automation News10-16ETH Zurich scientists turn WUJI robotic hand into self-propelled mobile manipulator
ETH Zurich researchers have developed a system where a WUJI robotic hand can crawl, recover from falls, and perform manipulation tasks without a separate mobility system. Using reinforcement learning, the hand learns to support its own weight and move using its fingers. The work was presented in a paper titled 'Fingers as Legs: Learning Self-Supported Locomotion and Manipulation with an Anthropomorphic Hand.'
Universal Robots — Official announcementsFirst-party10-21Teradyne Robotics and Sereact Partner to Scale Physical AI Solutions for Retail and Logistics
Teradyne Robotics and Sereact have announced a partnership to deploy AI-powered physical solutions in retail and logistics, using Universal Robots' arms with Sereact's Cortex platform. The collaboration includes live deployments at Zalando's fulfillment centers, demonstrating the ability to handle unstructured tasks like returns processing. The solution runs on NVIDIA technologies, including Isaac Sim and Jetson Thor, and is being scaled globally.
arXiv Robotics — research abstracts10-20EndoVLA-OOV: Language-Conditioned Out-of-View Target Acquisition in Robotic Endoscopy
This research presents EndoVLA-OOV, a vision-language-action policy for robotic endoscopy that translates anatomical descriptions into camera motions. The policy uses a pretrained backbone adapted to continuous distal bending commands and combines images to guide search. An open-source dataset with 1,050 episodes and 262k observation-action pairs supports policy learning. Physical evaluations show 93.3% success in cross-embodiment transfer, with success rates improving from 40.0% to 90.0% with more training data.
Motors Don't Need Upgrades, Robots Can Run Faster. Beihang University Science Subjournal Reveals New Logic for Quadruped Locomotion
Professor Shi Qing's team from Beijing Institute of Technology was inspired by the high-speed running mechanism of the elephant shrew. They designed a micro quadruped robot named FLEXOR equipped with a dual-joint coupled spine. Through dynamic spine-leg coordination, the robot achieved a 31.6% increase in speed and a 32.2% reduction in energy consumption without upgrading its motors. The study reveals the critical role of the timing coordination between the spine and legs in locomotion performance and verifies the universality of this mechanism across different robot morphologies, providing new insights for highly mobile embodied intelligent robots.
arXiv Robotics — research abstracts10-16Revisiting Embodied Chain-of-Thought for Generalizable Robot Manipulation
This research explores embodied chain-of-thought (CoT) for robotic control, introducing ERVLA, a vision-language-action model that effectively integrates linguistic reasoning into robot manipulation. ERVLA uses a CoT-dropout strategy during training and predicts actions directly during inference, achieving 86.9% success on LIBERO-Plus and 53.2% on VLABench, with superior performance in out-of-distribution settings.
arXiv Robotics — research abstracts10-20Refuse the Harm, Complete the Task via Graph-of-Rules Planning for LLM-Enabled Robots
This research presents RuleGraph, a multi-agent Graph-of-Rules planning framework that allows LLM-enabled robots to reject harmful parts of instructions while completing legitimate tasks. The method distributes safety enforcement across multiple reasoning stages, achieving 0% attack success rate and high task success rates on benchmark environments.
MIT — RoboticsFirst-party10-07Planning system ensures a robot’s flight path will remain collision-free
MIT researchers have developed a new autonomous navigation system called SANDO that guarantees collision-free flight for UAVs in dynamic, unknown environments. The system uses a time-sensitive safety corridor and mathematical proofs to ensure safety, even when obstacles are moving unpredictably. SANDO has been tested in simulations and real-world flights, successfully avoiding collisions in all scenarios.
NVIDIA Advances Isaac Platform for General-Purpose Robotics With New Reference Humanoid Frameworks, Simulation Pipelines and Data
NVIDIA announced a major expansion of the Isaac open robotics development platform, adding a reference humanoid, datasets, and runtime, along with updates to its robot learning framework. The Agile ONE SN humanoid, built on Isaac GR00T and Jetson Thor, is designed for research and development with integrated safety systems. The new technologies aim to accelerate the path from data to deployment for general-purpose robots.
arXiv Robotics — research abstracts10-28Rendering-Free Event Frame Simulation for Fast Reinforcement Learning of Autonomous Robot Navigation
This research proposes an analytic event simulator that generates event streams directly from scene geometry and ego-motion without visual rendering, enabling fast reinforcement learning for autonomous robot navigation. The simulator produces 64x64 event observations in 17 microseconds, achieving over 400x speedup compared to state-of-the-art rendering pipelines. The policy trained on this simulator transfers zero-shot to physically based simulations and achieves 83% success in unseen cluttered environments, with a 1.2 MB int8 network running onboard a 56-gram drone.
Robots & Hardware
Tech Xplore — Robotics10-30Robot Moth Grows UV Sensor Using Artificial Circulatory System
Researchers from the University of Tartu and the Italian Institute of Technology developed a moth-shaped robot that uses an artificial circulatory system to deliver chemical building blocks and grow a light sensor where UV light strikes. The robot's body is 3D-printed with microchannels that allow fluid to flow, enabling the synthesis of a conductive polymer sensor. This approach allows the robot to create new hardware on-site, enhancing its adaptability to unpredictable environments.
Industry
Robotics & Automation News10-30Bleu Robotics Raises €5 Million to Scale Humanoid Robot Deployment in Factories
Bleu Robotics, a Paris-based startup, has raised €5 million in a pre-seed round to scale its software solution for industrial humanoid robots. The company aims to reduce the full deployment process to under an hour by using abductive learning, which enables robots to learn new tasks from as few as ten operator demonstrations. This approach eliminates the need for large datasets and long training cycles, allowing for cost-effective and scalable deployment in manufacturing settings.