Skip to content

Technical Directions

Open artifacts Latest coverage

Robotics coverage relating to open artifacts.

309 selected reportsLast 30 days 306 recordsAll coverage 824 records

updated

Signal archive — Page 12

2026-11-02Mon221–240 records
  1. arXiv Robotics — research abstracts75

    Interaction-Aware Sampling-Based Motion Planning for Autonomous Surface Vessels in Urban Waterways

    This paper introduces a decentralized motion planner using model predictive path integral control for autonomous surface vessels in urban waterways. The method jointly plans trajectories for the ego vessel and predicts other vessels' paths to handle interactions, encoding traffic rules as discontinuous costs. Evaluated on real maps of Amsterdam canals with 780 randomized runs, it achieves collision-free goal reaching and outperforms optimization-based baselines. Physical validation on two scaled prototypes confirms onboard planning at 10 Hz successfully resolving head-on and crossing encounters without collisions or rule violations.

  2. arXiv Robotics — research abstracts75

    Direct Action-Manifold Modeling for Efficient and Robust Bimanual Manipulation

    This paper introduces the Direct Manifold Action Transformer (DMAT), a method that reformulates bimanual action generation as direct modeling of observation-conditioned joint-space action chunks to eliminate iterative inference latency. By anchoring predictions with local 3D geometry and applying State-Consistent Spectral-Time Refinement, DMAT achieves single-step inference while maintaining robustness under physical interventions. Evaluated across 567 standard and out-of-distribution trials, DMAT demonstrates superior aggregate success rates and faster generation speeds compared to Improved 3D Diffusion Policy (IDP3) and DiTFlow3D, establishing a new paradigm for efficient bimanual manipulation.

  3. arXiv Robotics — research abstracts75

    A complete analytical characterization of optimal collision-free motion for two robots

    This paper presents a complete analytical characterization of optimal collision-free trajectories for two robots starting and ending at rest under double-integrator dynamics with a squared-acceleration objective. The authors prove that these trajectories belong to eight distinct contact families, determine which family applies for each configuration of initial and final positions, analyze the geometrical properties of the optimal trajectories, provide a numerical algorithm for computing them, and discuss how these insights may improve the performance of existing numerical trajectory optimizers.

  4. arXiv Robotics — research abstracts75

    AC-SMPC: Differentiable Actor--Critic MPC with State Constraints for Safe and Fast Robot Navigation

    This research presents AC-SMPC, a novel actor-critic model predictive control framework designed for safe and efficient robot navigation in confined dynamic environments. The method addresses the challenge of integrating explicit state constraints into learning-based planning by representing robots and obstacles as vertices and half-spaces, locally linearizing nonlinear dynamics, and convexifying collision avoidance constraints into a differentiable MPC layer. An elastic quadratic program is formed using bounded slack variables to handle cases where local constraint approximations cannot be simultaneously satisfied. Simulation tests and experiments demonstrate the framework's ability to adaptively adjust learned policies for multi-robot navigation in narrow spaces while maintaining motion safety and improving efficiency over existing approaches.

  5. arXiv Robotics — research abstracts75

    Physically Constrained Kinematic Calibration for High-Precision Robotic Needle Guidance

    This paper presents a robust bound-constrained kinematic calibration method to improve positional accuracy for a 6-DoF robotic needle-guidance manipulator. By using physically plausible parameter ranges as active optimization constraints, the approach mitigates parameter coupling and overfitting. In simulation, the method reduced end-effector RMSE from 3.422 mm to 0.690 mm (79.8% improvement). Benchtop experiments achieved sub-millimeter validation accuracy, with 0.865 mm RMSE in trajectory validation and 0.610 mm RMSE in coordinate-lattice verification. These results establish a high-precision foundation for future clinical applications of robotic needle insertion.

  6. arXiv Robotics — research abstracts75

    DiffWBC: Unlocking Steerable General Humanoid Motion Tracking via One-Step Diffusion Whole-Body Control

    This paper introduces DiffWBC, a generative whole-body control framework for humanoid robots that unifies general motion tracking, test-time steering, and simulation-free failure discovery. The method utilizes a one-step diffusion model to achieve low local tracking error (20.1 mm on BONES) and zero-shot adaptation to various tasks including hurdle navigation and actuator faults without retraining or physics simulations.

  7. arXiv Robotics — research abstracts75

    HIPC: Humanoid Iterative Planning and Control via Goal-Conditioned Motion Closure

    This research introduces HIPC, a real-time framework for humanoid robots that unifies generative motion planning with closed-loop control. It addresses issues like spatial drift from open-loop paradigms and discontinuities from direct state feedback by using goal-conditioned planning, deviation-decayed recovery, history re-anchoring, and latency compensation. Experiments demonstrate that HIPC achieves state-of-the-art goal-reaching accuracy while maintaining robustness against physical disturbances and the ability to recover from falls.

  8. arXiv Robotics — research abstracts75

    Learning from Physical Experience: Context-Space Adaptation for Robotic Manipulation

    This research introduces LEXP, a framework that enables general-purpose reasoners like ChatGPT-6-Astra to perform physical manipulation tasks by adapting to context through experience. Across 12 real-world manipulation tasks, LEXP achieved 23 out of 36 successful trials (63.9%), significantly outperforming memory-only approaches (52.8%) and vanilla Astra (5.6%). The system combines task-level planning with an experience module that extracts lessons from execution evidence and provides contextual feedback for subsequent actions while keeping model weights and action interfaces fixed. Results show that both task-level planning and retained experience are critical components, as removing either reduces success rates substantially.

  9. Chinese robotics — Research news discovery75

    ManuPrax Robotics Announces Nearly $10 Million in Funding for Dexterous Robot Hand Systems

    On November 2, ManuPrax Robotics announced it has completed two consecutive funding rounds totaling nearly $10 million. Founded in September 2026 by Li Pu Hao, a Tsinghua University PhD student, and Chen Zeyuan, a Peking University master's graduate, the company focuses on developing dexterous hand systems with integrated software and hardware to enhance robot manipulation in unstructured environments. The rounds were led by Alphaist Partners and Yaotu Capital, with participation from an angel investor and a parts industry controller. Both founders previously conducted research at Beijing Institute for General Artificial Intelligence (BIGAI), where the company is also supported as an incubation project.

  11. Chinese robotics — Research news discovery75

    China Robotics Industry Development Alliance and AIROBO Sign Vietnam Order for Over 1,000 Robots

    On October 25, 2026, the China Robotics Industry Development Alliance (CDRA) and AIROBO signed a procurement agreement in Ho Chi Minh City, Vietnam, for over 1,000 robots. The order, directed at multiple large Vietnamese conglomerates, is reported by organizers as the largest single robot purchase from China to Vietnam. Delivery is scheduled for December 2026 across five application scenarios including education, security, commercial cleaning, cultural tourism performances, and delivery services.

  12. Chinese robotics — Robot learning research75

    Boyinhechuang Opens Suzhou Headquarters to Accelerate Industrial Embodied Intelligence

    On October 29, 2026, Boyinhechuang officially opened its new headquarters in Suzhou, positioning the facility as a central hub for R&D, simulation, validation, and delivery of industrial embodied intelligence solutions. The company, incubated by Bo Yuan Capital and Galbot, highlighted partnerships with United Automotive Electronics and others to deploy robots in real-world manufacturing scenarios. While specific production targets were outlined for future years, the immediate focus remains on integrating technology into complex factory environments.

  15. Robotics — Chinese web discovery75

    Embodied AI Robots Face Real-World 'Final Exam' in Shanghai

    The 2026 Global Developer Pioneer Conference and International Embodied AI Skills Competition was held in Shanghai, featuring a rigorous real-world testing environment with nine major tracks and 23 competition items. The event highlighted the gap between simulated and real-world performance, where robots trained on billions of yuan failed to perform effectively in actual home settings. Key scenarios included industrial inspection tasks like navigating complex terrain and reading gauges within 30 minutes, and retail operations requiring stable manipulation of soft-packaged goods without spilling. Post-event, a 2,000-square-meter space was permanently converted into a training and validation platform for embodied AI robots, focusing on manufacturing, commercial services, and public utility applications.

  16. Robotics — Chinese web discovery75

    Amicro Technology (2065.HK) Files for IPO as Robot Chip Platform Sees Growth Opportunity

    Amicro Technology, a provider of dedicated SoCs for robot SLAM, navigation, and mapping, is filing for an IPO on the Hong Kong Stock Exchange under Chapter 18C rules. The company's proprietary Amicro Core platform integrates self-developed algorithms with chips, sensors, and motion control modules. Revenue grew from approximately 206 million yuan in 2023 to 407 million yuan in 2025, with multi-modal solutions becoming the primary revenue source by 2025.

  17. Chinese robotics — Company news discovery75

    Meta Partners with Fourier Intelligence to Develop Humanoid Robot 'Max' Amid Physical AI Competition

    According to a November 2, 2026 report by Huaxin Securities, Meta has entered a deep cooperation with Chinese humanoid robot company Fourier Intelligence to develop its first humanoid robot, 'Max'. The prototype is positioned as the physical carrier of the Meta Muse large model for home chores. Development began in September of the current year, with a second prototype already underway.

  18. Chinese robotics — Research news discovery75

    Chinese Academy of Sciences Researchers Develop Biomimetic Adaptive Electronic Skin for Robot Tactile Perception

    A team led by Researcher Wei Dapeng at the Chongqing Institute of the Chinese Academy of Sciences has developed a flexible electronic skin based on a biomimetic adaptive mechanism. Published in Nano Research, the device uses a magnetic conductive network to dynamically adjust sensitivity and signal-to-noise ratio under external stimuli. The sensor operates across a wide pressure range from 1 Pa to 3.5 MPa, achieving high sensitivity and fast response, offering new possibilities for robot intelligence in complex environments.

  19. Chinese robotics — Company news discovery75

    Chengdu-made robots move from exhibition stunts to real-world shifts: emotional interaction models launch at 98k yuan, cultural tourism guides serve over 100k visitors, and humanoid output exceeds 40k

    Sichuan Embodied Technology secured over 100 million yuan in Pre-A funding for its 'Ai Qiu' emotional interaction robot and production scaling. The Chengdu Natural History Museum features the 'Xiao Xin' cultural tourism guide robot, which has served over 100,000 visitors. Ministry of Industry and Information Technology data indicates 2026 humanoid robot production exceeded 40,000 units with a forecast to surpass 100,000 annually. Chengdu's 'Jiaozi Avenue' district hosts deployed delivery and inspection robots facing real-world traffic and weather conditions. Local companies like Yuefan Innovation and RoboCity are refining urban logistics and governance applications. The city aims for a 50 billion yuan embodied AI industry scale by 2027 under new government action plans.

2026-11-01Sun
  1. Chinese robotics — Company news discovery75

    Huiyi Intelligence Launches A-Share IPO Push as Robot Chip Maker

    On October 29, Huiyi Intelligence officially initiated its public offering campaign on the A-share market. The company, which positions itself as an early entrant in embodied intelligence computing platforms, has achieved scaled commercial deployment of its robot chips and domain controllers. Its development path reflects a trend of autonomous driving technology migrating to robotics, signaling the industry's shift from technical validation to commercial verification.

Timezone · UTC

Article dates follow your selected timezone. Briefing editions use Hong Kong time (UTC+8).