Industry Briefing
A single destination for timely, editor-curated robotics news from around the world.
Gravity Develops Unified Framework for Long-Range Complex Robotic Tasks
Gravity has introduced a unified embodied intelligence framework designed for long-range and complex robotic tasks. This framework, built on a Mixture-of-Transformers (MoT) architecture, integrates visual language models (VLM) for instruction and scene understanding, task reasoning, and world modeling to predict future states and evaluate sub-goals. It also incorporates tactile and force feedback, prior knowledge, and multi-modal supervision to enhance task execution and adaptability. The significance of Gravity's framework lies in its ability to improve the success rate of complex operations that require precise contact and autonomous error correction. By combining AR Transformer and Diffusion Transformer, Gravity enables robots to simulate multiple strategies and assess risks before executing tasks. This advancement shifts robotic capabilities from reactive responses to proactive planning, making it suitable for applications in precision assembly, complex sorting, and flexible manufacturing. Looking ahead, Gravity aims to further develop its complete system, having already implemented components like Gravity VLA and Gravity 4D WAM. The focus will be on enhancing the framework's ability to learn from real-world experiences, thereby creating a continuous feedback loop that improves operational efficiency and adaptability in various industrial contexts. No further timeline was disclosed at the time of publication.
Robotic Frameworks Embodied Intelligence Task Automation Machine Learning Robotics
Ensuring Connectivity for Humanoid Robots with Advanced 6G Technology Solutions
At the WAIC 2026, 208 embodied intelligent terminals and over 300 real machines showcased their capabilities, raising concerns about connectivity during real-time collaborative operations. Asmote introduced a next-generation wireless base designed for humanoid robots, which integrates communication, perception, intelligent computing, and control based on 6G technology. This innovation is crucial as it addresses the need for reliable connectivity in complex industrial environments, ensuring that robots remain online with 99.99% reliability and millisecond-level latency. Asmote's AI-EDGE platform significantly reduces the power consumption, cost, and weight of robots by offloading communication and computing tasks to the 6G-native base, allowing robots to focus on performance optimization and task execution. Looking ahead, Asmote aims to provide a digital foundation for physical AI across various industries, emphasizing the importance of a robust infrastructure that goes beyond mere connectivity. No further timeline was disclosed at the time of publication.
Humanoid Robots 6G Technology Wireless Communication AI Industrial Automation
Launch of the 2026 Intelligent Embodiment Forum at WAIC Focuses on Physical AI
On July 19, 2026, the Intelligent Embodiment Forum was successfully launched at the World Artificial Intelligence Conference (WAIC). This event, co-hosted by Zhiyuan and Mifeng Technology, gathered leading experts from top universities and tech companies to discuss the evolution of physical AI and its challenges, including data scarcity and the need for unified physical representation. The forum highlighted the importance of transitioning from digital AI to physical AI, emphasizing the need for a comprehensive understanding of the environment and task execution. Keynote speaker Yao Maoqing outlined the three critical barriers to scaling physical intelligence: data, representation, and feedback loops. He stressed that overcoming these barriers is essential for the development of general embodied intelligent systems. Looking ahead, the forum set the stage for ongoing collaboration between academia and industry to accelerate the evolution of physical AI. No further timeline was disclosed at the time of publication.
Physical AI Robotics Artificial Intelligence Machine Learning
WAIC 2026: Insight AI Unveils First Embodied Semantic Intelligence System, insightOS Semantic
At the 2026 World Artificial Intelligence Conference in Shanghai, Insight AI launched the world's first embodied semantic intelligence system, insightOS Semantic. This system represents a pivotal shift in the embodied intelligence industry, moving from mere demonstration capabilities to practical applications in real-world scenarios. The significance of this development lies in its comprehensive approach to embodied intelligence, which now requires not only technical prowess but also a robust operating system, scenario validation, and a thriving developer ecosystem. Insight AI aims to address the critical challenges of understanding, adaptability, and evolution in robotics, which have hindered the large-scale deployment of embodied systems. Looking ahead, Insight AI's insightOS Semantic is designed to facilitate seamless communication between humans and robots, enabling task execution through natural language. The system's architecture integrates semantic understanding with physical operation capabilities, promising to enhance the efficiency and intelligence of robots in dynamic environments. No further timeline was disclosed at the time of publication.
Embodied Intelligence Robotic Systems AI Technology Natural Language Processing
Fourier Introduces 'Embodied Home' Solution for Intelligent Robot Operations in 2026
In 2026, the embodied intelligence industry is transitioning from technology validation to large-scale commercialization. With support from capital and policy, the focus is shifting from rigid, high ROI scenarios like industrial manufacturing to commercial services and home environments. A key challenge remains: can robots effectively operate in the complex and unstructured home settings? The complexity of home environments poses significant challenges for humanoid robots, which must navigate unclear instructions and varied layouts. Current mainstream navigation AI struggles with personalized understanding, a critical shortcoming for practical applications in home services. At the World Artificial Intelligence Conference 2026, Fourier showcased the 'Embodied Home' solution, aiming to enable robots to understand their environment and intentions, thereby completing long sequences of tasks autonomously. The 'Embodied Home' solution integrates a semantic task execution hub for humanoid robots, combining large language models, 3D spatial semantic memory, navigation planning, and control execution. This allows robots to autonomously interpret and execute tasks based on natural language commands, enhancing user interaction and ensuring task reliability through a complex task scheduling mechanism. No further timeline was disclosed at the time of publication.
Robotics AI Home Automation Semantic Understanding
Efort to Showcase Universal Base Technology at WAIC 2026 for Robotics Advancement
Efort's QiZhi will present its Universal Base technology at the WAIC 2026, aiming to enhance the capabilities of robots. The technology focuses on providing a foundational infrastructure that allows robots to learn and perform tasks across various scenarios, addressing the industry's need for a generalized capability rather than just hardware improvements. This initiative is crucial as the robotics sector has seen a significant influx of capital, with funding in the domestic embodied intelligence sector surpassing 93.5 billion yuan in the first half of 2026. However, the industry consensus highlights a gap in data availability and the need for a robust framework to enable effective learning and task execution by robots. Looking ahead, Efort's approach could redefine the landscape of robotics by shifting the focus from hardware specifications to a comprehensive technological foundation. As the industry transitions from demonstration to practical applications, the success of this Universal Base technology will be pivotal in overcoming existing challenges and enhancing the deployment of robots in real-world scenarios. No further timeline was disclosed at the time of publication.
Robotics AI Industrial Automation Machine Learning
StepFun Launches STEPX Neo: An AI Agent Phone with Cross-App Integration
StepFun, a Shanghai-based AI unicorn, has introduced the STEPX Neo, an innovative phone equipped with the agent-native Step AOS and Amoo agent. This device integrates over 10 major platforms, including Alipay, Meituan, Didi, and Baidu, enabling autonomous task execution across applications. The launch of STEPX Neo is significant as it represents a leap forward in mobile technology, allowing users to perform tasks seamlessly across multiple applications without manual intervention. By leveraging the capabilities of various platforms, the STEPX Neo aims to enhance user experience and efficiency in daily tasks. Looking ahead, the impact of STEPX Neo on the market will be worth monitoring, particularly in how it influences user adoption of AI-driven mobile solutions. No further timeline was disclosed at the time of publication.
Technology
LLMs help robots understand vague instructions and focus on key details
Researchers at MIT have developed an innovative approach to enhance the efficiency of robots in performing chores in various environments, including homes and factories. This new method employs a dual-language model system: the first model is designed to interpret and clarify user instructions, while the second model focuses on filtering out irrelevant information that may hinder task execution. This advancement aims to improve the interaction between humans and robots, making it easier for machines to understand and carry out complex tasks effectively. The initiative reflects MIT's commitment to advancing robotics technology and its potential applications in everyday life.
School of Engineering MIT Schwarzman College of Computing Aeronautical and astronautical engineering Electrical engineering and computer science (EECS) Computer Science and Artificial Intelligence Laboratory (CSAIL) Computer science and technology
Learning flight navigation like a honey bee
In June 2026, a groundbreaking study published in Science Robotics highlights significant advancements in robotic technology aimed at enhancing human-robot collaboration. Researchers from leading universities and tech companies have developed a new generation of robots equipped with advanced artificial intelligence, enabling them to perform complex tasks alongside human workers more efficiently. This initiative is driven by the increasing demand for automation in various industries, particularly in manufacturing and healthcare, where precision and reliability are paramount. The study outlines how these robots can adapt to dynamic environments, learn from human interactions, and improve their performance over time, thereby reducing the risk of workplace accidents and increasing productivity. The research team conducted extensive field tests in factories and hospitals to evaluate the robots' capabilities in real-world scenarios. The results demonstrated a marked improvement in task execution speed and accuracy when robots and humans worked in tandem, showcasing the potential for these technologies to transform traditional workflows. As industries continue to evolve, the implications of this research could lead to a new era of collaboration between humans and machines, addressing labor shortages and enhancing operational efficiency. The findings underscore the importance of ongoing innovation in robotics, paving the way for future developments that could redefine the nature of work across various sectors.
Editors' Choice
Cobot Announces Second-Generation Proxie, Bringing Production-Tested Physical AI to Real Operations
A new production-tested platform has been developed that integrates mobility, bimanual manipulation, and self-directed task execution, allowing for rapid deployment within days. This innovative system operates independently, requiring no IT integration or human dispatcher to function effectively. The platform aims to streamline operations and enhance efficiency in various settings, making it a significant advancement in automation technology. With its user-friendly design, it is poised to transform workflows across industries by minimizing the need for extensive setup and oversight.
One Day vs. Seven Days: Post-Training Revolutionizes Embodied Intelligence Deployment Speed
At the ICRA 2026 REAL-I Challenge, the NUS-CLEAR team achieved a remarkable milestone by completing intricate robotic tasks in a single day, a process that typically required engineers between five to seven days. This significant improvement in efficiency is largely due to their pioneering post-training and real-time evaluation system, which optimizes model performance and adaptability for real-world applications. As the robotics industry increasingly prioritizes deployment efficiency over mere model capability, the development of standardized post-training tools is becoming crucial for enabling rapid task execution.
Embodied Intelligence Post-Training Robotics AI Deployment Automation
TARS DexHand Makes Global Debut at ICRA: Redefining Dexterous Operations with Mind-Hand Integration
At the ICRA conference held in Vienna, TARS DexHand was unveiled, showcasing an innovative 'mind-hand' integration system designed to mimic human hand movements with 21 degrees of freedom. This advanced technology enhances task execution by utilizing sophisticated data alignment and modeling capabilities. During the demonstration, the system performed real-time interactions and complex tasks, underscoring its significant potential for industrial applications and the advancement of embodied intelligence. The event highlighted the ongoing evolution in robotics and its implications for various sectors.
Dexterous Robotics Embodied Intelligence Robotics Automation AI Technology
X-Square Robot Unveils WALL-WM, the World's First Event-Level Prediction Embodied AI World Model
X-Square Robot has unveiled WALL-WM, a groundbreaking AI world model that represents a significant advancement in event-level prediction. This innovative technology transitions from traditional frame-by-frame prediction to a more sophisticated semantic event understanding, allowing robots to comprehend task objectives rather than merely memorizing sequences of pixels. The launch of WALL-WM marks a pivotal moment in robotics, as it enhances the ability of machines to interpret and interact with their environments more intelligently. This development is expected to revolutionize various applications in robotics, providing a more intuitive and effective approach to task execution.
EmbodiedAI
Tencent launches WorkBuddy productivity AI agent for global users
Tencent Cloud has introduced WorkBuddy, a new AI productivity agent designed for users worldwide, following its initial launch in China. The platform aims to enhance office workflows by utilizing natural language prompts to decompose tasks, integrate with external tools, and produce deliverables for various work and study environments. WorkBuddy also facilitates remote task execution, making it a versatile tool for both professionals and students. This launch reflects Tencent Cloud's commitment to improving efficiency and collaboration in an increasingly digital workspace.
News Feed
Revolutionizing Robot Interaction in the Era of Physical AI with Agile Robots
Agile Robots has unveiled a groundbreaking force control technology aimed at revolutionizing industrial automation and artificial intelligence. This innovative development, rooted in decades of aerospace research, allows robots to dynamically adapt to their physical environments, effectively addressing the persistent 'last millimeter problem' that has hindered task execution in complex scenarios. By enhancing the reliability of physical interactions, Agile Robots is positioning itself at the forefront of the evolving landscape of automation, enabling more sophisticated and efficient operations in various industries. This advancement marks a significant step forward in the integration of robotics into real-world applications, promising to improve productivity and operational effectiveness.
Force Control Technology Industrial Automation Robotics AI Physical Interaction
Breakthrough! Launch of the Genie Operator-2 Model: Embodied Intelligence Overcomes the Key Barrier of 'Knowing and Doing'!
On April 6, Zhiyuan Robotics introduced the Genie Operator-2 (GO-2), an advanced embodied intelligence model aimed at improving the connection between semantic understanding and physical execution in robotics. The new model features a cutting-edge architecture and has been developed using extensive training data, which collectively contribute to its enhanced performance in various robotic benchmarks. This innovation is expected to significantly increase the reliability of task execution in robotic applications, marking a notable advancement in the field of robotics.
Embodied Intelligence Robotics AI Automation
How to teach the same skill to different robots
In modern manufacturing, the process of upgrading a fleet of robots frequently requires a complete overhaul, involving both hardware replacement and extensive reprogramming of tasks. This challenge arises even when robots are designed for similar functions, as variations in joint configurations and movement restrictions can complicate task execution. As industries strive for efficiency and adaptability, addressing these complexities is crucial for optimizing robotic performance and maintaining productivity in dynamic production environments.
ZhiYuan Research Institute Launches RoboClaw Embodied Intelligence Operating System for Plug-and-Play Robots
The RoboClaw operating system, a collaborative effort by ZhiYuan Research Institute, Xuanji Intelligent, and Mianbi Intelligent, has been unveiled as a groundbreaking solution for robotic automation. This innovative system allows robots to perform tasks autonomously through natural language commands, removing the necessity for coding or specialized operational skills. By facilitating seamless communication between artificial intelligence and physical task execution, RoboClaw aims to enhance the efficiency and accessibility of robotic technology. This development represents a significant step forward in making advanced robotics more user-friendly and versatile, potentially transforming various industries that rely on automation.
Embodied Intelligence Robotics AI Automation SaaS
Tesla Optimus Skills Packages (2026): What It Can Do
A comprehensive skills catalog has been established, detailing confirmed factory tasks and home skills, distinguishing between proven capabilities and those demonstrated only in theory. This extensive catalog encompasses over 3,000 tasks, providing a structured framework for skill assessment and development. The initiative aims to enhance training and operational efficiency within various sectors. With data compiled up to October 2023, this catalog serves as a vital resource for organizations seeking to optimize workforce skills and ensure effective task execution.
Understanding Robotics and Autonomous Systems in Next-Gen Smart Factories
Factories are experiencing a significant transformation as they shift from traditional linear automation to interconnected, intelligent production ecosystems. This change is driven by the integration of robotics and autonomous systems, with JAKA positioning its industrial cobots as essential components within this advanced network. The industrial cobot serves as a flexible link between manual workstations and fully automated lines, enabling manufacturers to adapt quickly to varying tasks such as assembly and inspection. This adaptability is crucial for high-mix production environments, allowing workflows to respond in near real-time to digital directives from a central Manufacturing Execution System (MES). The evolution of true autonomy is facilitated by the connectivity of individual machines within a coordinated network. JAKA's robotic arms can communicate seamlessly with autonomous mobile robots and vision systems, creating a responsive operational loop. For example, an autonomous robot can deliver components to a JAKA cobot, which then executes tasks based on cloud-based analytics. JAKA emphasizes the importance of reliable, connected hardware to support these systems, focusing on high-precision control technology that ensures accurate task execution. This reliability is essential for advancing towards lights-out production in certain processes. As the next-generation smart factory emerges, the industrial cobot is positioned as a versatile agent within a complex architecture of robotics and autonomous systems. JAKA aims to provide manufacturers with the necessary hardware to build increasingly responsive and autonomous production environments, enhancing efficiency and complexity management in the manufacturing landscape.
Teaching robot policies without new demonstrations: interview with Jiahui Zhang and Jesse Zhang
At the Conference on Robot Learning (CoRL) 2025, researchers Jiahui Zhang, Yusen Luo, Abrar Anwar, and Sumedh A. Sontakke introduced the ReWiND method, a novel approach designed to enhance robotic learning through language-guided rewards. This method unfolds in three distinct phases: first, it involves learning a reward function; next, it incorporates pre-training; and finally, it applies the learned reward function alongside the pre-trained policy to tackle new language-specific tasks in real-time. The motivation behind this research is to enable robots to adapt to new tasks without requiring additional demonstrations, thereby streamlining the learning process. By leveraging language as a guiding tool, the ReWiND method aims to improve the efficiency and effectiveness of robotic task execution.
AgiBot unveils Lingxi X2, an advanced humanoid robot with multimodal intelligence
Peng Zhihui, the founder of AgiBot, unveiled a video today featuring the Lingxi X2, an advanced bipedal humanoid robot. This innovative robot is equipped with a sophisticated computing engine that enables it to perform a variety of tasks, including walking, running, dancing, riding a bicycle, balancing on a hoverboard, and even reading medication instructions. The Lingxi X2 was developed by AgiBot's X-Lab, highlighting the company's commitment to advancing robotics technology. This release marks a significant step in the integration of motion, interaction, and task execution intelligence in humanoid robots, showcasing the potential for practical applications in everyday life.
News Feed
Domestic AI Unicorn Unveils MiniCPM-Robot to Enhance Robot Memory at WAIC
At the World Artificial Intelligence Conference (WAIC), a notable trend emerged in the robotics sector, with an increasing number of VLA models attempting to enable robots to execute longer task chains. A significant challenge has been the fundamental conflict between 'contextual memory' and 'real-time inference costs' in VLA models, which has hindered the completion of long-range tasks. During WAIC, Mianbi Intelligence introduced and open-sourced its first series of embodied intelligence results, the MiniCPM-Robot, which includes the general VLA model MiniCPM-RobotManip and the tracking navigation model MiniCPM-RobotTrack. This model, with only 1.3 billion parameters, successfully completed long-range tasks like making sandwiches and achieved a significant lead over well-known models in the RMBench leaderboard, showcasing its advanced contextual memory capabilities. The release of MiniCPM-Robot signifies Mianbi Intelligence's transition of its accumulated multimodal technology from the digital realm to the physical world. The MiniCPM-Robot series addresses a structural shortcoming in current embodied intelligence, allowing for efficient visual token compression that enhances real-time inference speed while retaining memory, thus overcoming a long-standing dilemma in VLA model development.
Robotics AI Models Task Execution Memory Retention
RealBOT Robot Excels in Culinary Tasks as a 'Xinjiang Chef
On July 15, 2026, the World Humanoid Robot Sports Championship's domestic third stop launched simultaneously in Yining, Xinjiang, and Sai Lake, Bozhou. RealBOT, a wheeled humanoid robot by Ruierman, previously provided logistical support at the 2026 Chinese Dragon Boat Race in Bazhong, Sichuan. In this event, the robot transitioned from a service role to that of a 'Xinjiang chef' in a real kitchen environment, successfully completing intricate tasks such as noodle pulling, meat skewering, grilling, and vegetable cutting. This demonstration is significant as it showcases the robot's ability to adapt to unpredictable physical variables in real-world scenarios, aligning with Ruierman's philosophy that robots must be deployed in authentic environments to truly understand and perform tasks. The challenges faced by the robot, including precise force control and real-time tactile feedback, highlight the advanced capabilities of Ruierman's technology, including a mechanical arm with ±0.05mm repeatability and millisecond-level force feedback. Looking ahead, the successful execution of these culinary tasks emphasizes the importance of robust hardware and data capabilities in robotics. No further timeline was disclosed at the time of publication, but the ongoing development of integrated joint modules and high operational reliability positions Ruierman to address industry challenges related to traditional robots' limitations in various environments.
Humanoid Robots Culinary Robotics AI Automation Technology
Stardust AI Launches Lumo-2: Innovative Robot Action Model for Home Automation
On July 15, Stardust AI introduced its second-generation embodied base model, Lumo-2, which is the industry's first household latent world-action model. This launch includes the physical AI symbiotic agent, Agent Philia, enhancing their full-stack architecture of AI models, embodied operating systems, and rope-driven entities. The company will showcase its 'trinity' multi-scenario implementation solutions at the World Artificial Intelligence Conference in Shanghai from July 17 to 20. Lumo-2 autonomously performs 22 complex household tasks, demonstrating industry-leading capabilities in task range and complexity. This model addresses the challenges faced by robots in open environments, such as the inability to explain actions and the high costs of training complex skills. By predicting future scenarios before generating actions, Lumo-2 aims to overcome these bottlenecks and improve the practical execution of robotic tasks. Looking ahead, Stardust AI plans to enhance the scalability of Lumo-2 by expanding training data diversity and exploring efficient data engineering paradigms. The team is also focused on advancing real-world interactive learning to enable robots to adapt and evolve autonomously in dynamic environments. No further timeline was disclosed at the time of publication.
Household Robotics Physical AI AI Models Robotic Automation
Microsoft announces "Surface Laptop Ultra," designed with NVIDIA, enabling local execution of AI models.
Microsoft has unveiled its latest notebook, the Surface Laptop Ultra, marking a significant advancement in its product line. This new device is the first Surface model to feature the RTX Spark system-on-chip (SoC), which was co-designed with NVIDIA. The Surface Laptop Ultra boasts an impressive AI computing performance of 1 petaflop, allowing it to handle complex tasks efficiently. It is equipped with up to 128GB of unified memory, enabling the execution of models with 120 billion parameters locally. The highly anticipated laptop is set to be released in the fall of 2026, reflecting Microsoft's commitment to integrating cutting-edge technology into its devices to meet the growing demands of AI applications.
Shenzhen's Robotics Sector Advances 'Intermediate Layer' for Humanoid Robots at WAIC
At this year's WAIC, Shenzhen's robotics companies focused on enhancing the 'intermediate layer' necessary for humanoid robots to translate thoughts into actions. This layer is crucial for bridging the gap between model capabilities and real-world execution, addressing the industry's shift from 'can it be done' to 'can it be used'. The importance of this development lies in the need for robots to effectively process diverse training data sources, such as human demonstrations and simulation synthesis. Companies like Cross-Dimensional Intelligence introduced the Aligned DexWorld, which aligns data in a unified 3D physical space, while the DexVerse simulation engine generates varied training materials. Demonstrations showcased robots performing complex tasks, emphasizing the necessity for continuous action planning and adjustment in real-time. Looking ahead, the challenges remain regarding the amount of data required for new tasks and the stability of long-term operations. However, Shenzhen's robotics sector is clearly on the right path, focusing on completing the 'intermediate layer' to enable robots to transition from understanding to executing tasks effectively.
Humanoid Robots Robotics Technology AI Automation Data Alignment
Zhejiang Humanoid NAVIAI Advances Embodied Intelligence with SPIRE and EvoStack Systems
In 2026, the robotics industry is shifting focus from basic functionality to achieving results in real-world applications. Companies must now demonstrate their ability to translate genuine needs into systemic capabilities, ensuring robots can operate effectively in various environments. This transition marks a significant evolution in the industry's landscape. Zhejiang Humanoid has positioned itself uniquely by addressing real demands early on, integrating technology, talent, and industry chains to create a robust 'industrial carrier.' The company emphasizes that the physical robot is just the entry point, while the systemic capabilities form the core competitive advantage, enabling robots to adapt and perform in diverse scenarios. The NAVIAI product matrix includes advanced bipedal robots and versatile robotic arms designed for complex tasks. The SPIRE technology framework enhances robot performance by bridging the gap between understanding and execution, while the EvoStack toolchain supports continuous adaptation and improvement. As the industry evolves, the ability to navigate uncharted territories will define the leaders in robotics.
Humanoid Robots Embodied Intelligence Industrial Automation Robotics Technology
PNDbotics Unveils Full-Stack Development at WAIC 2026 with Adam Robots
On July 17, 2026, the World Artificial Intelligence Conference (WAIC 2026) commenced in Shanghai, featuring over 1,100 global exhibitors. PNDbotics showcased its full-stack development capabilities with the Adam full-sized humanoid robot and the Adam-U upper-body robot. Demonstrations included stair climbing, motion capture, and remote operation, highlighting the technology's robustness in unstructured environments. The significance of PNDbotics' presentation lies in its comprehensive display of technology from data collection to physical execution. The Adam robot, standing at 167 cm with 43 degrees of freedom, performed a live demonstration of stair climbing, marking its first public showcase in such a scenario. Additionally, the Adam-U robot demonstrated its adaptability by executing tasks based on natural language commands, showcasing the potential for advanced human-robot interaction. Looking ahead, PNDbotics is accelerating mass production following significant funding received in February 2026. The company is positioned to address industry challenges in embodied intelligence data scarcity while continuing to innovate in motion control and deep reinforcement learning. No further timeline was disclosed at the time of publication.
Humanoid Robots AI Robotics Technology Motion Control Remote Operation
Six Robots Assemble a 3.5-Meter Great Wall Model Using 81,920 Blocks at WAIC
At the World Artificial Intelligence Conference (WAIC), six robots from General Embodied Intelligence Company, Yuanli Lingji, undertook a challenging task to assemble a Great Wall model using 81,920 micro building blocks over 15 hours. Each robot was required to complete approximately 910 assembly actions per hour, achieving a speed comparable to skilled human workers. This demonstration highlights the complexities of robotic assembly, as traditional industrial robots operate under fixed conditions, while block assembly requires real-time perception and adaptability to varying positions and angles. The robots needed to maintain sub-millimeter precision throughout the task, pushing the limits of robotic capabilities and mimicking human dexterity. The execution team consisted of four desktop robots and two humanoid wheeled robots, each equipped with independent perception, decision-making, and execution abilities. The challenge tested multi-agent collaboration in a dynamic environment, emphasizing the need for real-time negotiation and coordination among robots to adapt to unforeseen circumstances during the assembly process. No further timeline was disclosed at the time of publication.
Robotic Assembly AI Technology Collaborative Robotics Precision Engineering
Robbyant Launches LingBot-VA 2.0, the First Embodied-Native AI Model for Robotics
Robbyant, a company under China's Ant Group, has introduced LingBot-VA 2.0, touted as the first embodied-native video-action world model specifically designed for robotics. Unlike traditional models adapted from digital content, LingBot-VA 2.0 is built from the ground up for physical-world tasks, enhancing physical accuracy and execution efficiency through its autoregressive architecture. This innovation is significant as it marks a departure from conventional robotics models that often compromise real-world performance by relying on video generation systems. Robbyant's approach allows for better prediction of how robot actions affect their environment, thus improving generalization and operational effectiveness in real-world applications. Looking ahead, Robbyant's LingBot-VA 2.0 is expected to advance the capabilities of robots in various tasks, demonstrated through its performance in complex scenarios such as preparing breakfast and unpacking deliveries. No further timeline was disclosed at the time of publication.
AI and Robotics
Yuanluo Technology Unveils First Autonomous Laboratory Utilizing Object-Centric Physics Model
Yuanluo Technology has successfully launched the world's first autonomous laboratory on a national research platform, marking a significant advancement in embodied intelligence. The laboratory's robotic system can autonomously perform over 40 operations, including nucleic acid extraction and cytotoxicity testing, with a precision of less than one millimeter. This achievement demonstrates the robot's capability to execute complex, multi-step tasks continuously for over three hours, addressing challenges in throughput and consistency in biochemical research. This development is crucial as it signifies a shift from demonstration to practical application of embodied intelligence in the biochemical and material science sectors. The Object-centric Physics Native Model (OPN), developed by Yuanluo, enables the robot to understand and adapt to the dynamic conditions of a real laboratory environment. By integrating visual, tactile, and force feedback, the robot can make real-time adjustments, ensuring stable execution of intricate experimental workflows across multiple devices. Looking ahead, the successful implementation of this autonomous laboratory sets the stage for further advancements in research and development processes across various industries, including public health and advanced manufacturing. The next milestones will involve expanding the capabilities of the OPN model and integrating it into more complex industrial systems. No further timeline was disclosed at the time of publication.
Autonomous Laboratories Embodied Intelligence Biochemical Research Robotics AI
World’s first surgery using teleoperated humanoid robots conducted by US team
A team of researchers has made a groundbreaking advancement by successfully employing teleoperated humanoid robots to carry out two complex tasks. This achievement marks a significant milestone in robotics, showcasing the potential for remote-controlled machines to perform intricate functions that were previously thought to be the domain of human operators. The experiments took place in a controlled environment, allowing the researchers to meticulously assess the robots' capabilities and refine their performance. The motivation behind this innovative project stems from the desire to enhance efficiency and safety in various industries, particularly in situations where human presence may be risky or impractical. By leveraging advanced teleoperation technology, the researchers demonstrated that humanoid robots could effectively mimic human movements and decision-making processes, paving the way for future applications in fields such as disaster response, healthcare, and manufacturing. The successful execution of these tasks not only highlights the progress in robotic technology but also opens new avenues for research and development, potentially transforming how we approach complex operations in challenging environments. As the team continues to refine their methods, the implications of this work could lead to widespread adoption of teleoperated robots in various sectors, ultimately improving productivity and safety for human workers.
AI and Robotics
Palladyne AI Executes $4.2 Million U.S. Air Force Contract to Advance Swarming Capabilities for Integrated Cross-Domain Operations
Palladyne AI Executes $4.2 Million U.S. Air Force Contract to Advance Swarming Capabilities for Integrated Cross-Domain Operations Visit http://www.palladyneai.com for further information Palladyne AI’s SwarmOS™ platform to support satellite integration, marking a major expansion of its multi-domain autonomy and ISR capabilities across space, air, maritime, and land 07/07/26, 06:15 AM | Mobile Robots, Other Topics | Palladyne AI Corp. Palladyne AI Corp. (NASDAQ: PDYN and PDYNW) ("Palladyne AI"), a developer of artificial intelligence software for robotic platforms in the defense and commercial sectors, today announced that it has executed the previously announced contract awarded by the Air Force Research Laboratory (AFRL) to solve one of the most persistent challenges in modern defense operations—how to make different autonomous systems work together as one coordinated team. The "Hierarchical Adaptive Networked Game-Theoretic Integration of Multiple Echelons (HANGTIME)" contract will address this need. More Headlines A3's Automate 2026 Breaks Records as Demand for Robotics, AI and Automation Grows NVIDIA and Hugging Face Bring New Models and Frameworks to LeRobot for the Open Robotics Community ABB Robotics completes its AI-powered Visual SLAM AMR portfolio with new autonomous forklift UMA Unveils Its Vision for the Next Generation of Humanoid Robots Robbyant Unveils LingBot-Depth 2.0 and LingBot-Vision to Redefine Robotic Spatial Perception Articles Unleash AI Innovation: The Power of NVIDIA RTX PRO 6000 Blackwell Workstation Edition Fueled by PNY-Supplied GPUs Automate 2026 Q&A with DESTACO Automate 2026 Q&A with Roboteon Advances in Robots to See & Interpret within Warehouse Environments Building Resilient Fulfillment Networks with Robotics and Real-Time Logistics Data Today, drones, ships, and satellites often operate largely independently, limiting how quickly warfighters can see and respond to threats. HANGTIME will utilize Palladyne AI's patented SwarmOS™ software platform—the defense variant of the Palladyne™ Pilot embodied AI software—as the baseline technology to bridge that gap, connecting disparate systems so they can share intelligence, adapt to changing conditions, and act in sync across domains, including space, air, maritime, and land. By integrating satellites for the first time, this project also extends Palladyne AI's technology from the ground to orbit, enabling faster, more informed decision-making and coordinated mission execution, turning tactical commanders into strategic commanders by giving them more cross-domain intelligence, surveillance, and reconnaissance (ISR) capabilities than ever before. "Our collaboration with AFRL showcases what's next for autonomous operations," said Ben Wolff, President and CEO, Palladyne AI. "This isn't about replacing humans—it's about giving them sharper, faster insight. By connecting satellite, aerial, and ground systems using the patented SwarmOS embodied AI platform as a foundational technology, we're helping the warfighter make better decisions in real time and stay one step ahead on the battlefield." "The HANGTIME project is a breakthrough that unites high-altitude assets and situational unmanned systems into one coordinated sensor network—delivering a major advantage for the defense industry," said Dr. Denis Garagic, Chief Technology Officer, Palladyne AI. "For the first time, a single AI framework can coordinate assets across multiple domains, including satellites. That means these systems can now think and act together as a team, sharing what they see and learning as conditions change." "The HANGTIME effort represents a critical step in multi-domain autonomy for coordinated execution in challenging environments," said Caleb Williams, Program Manager, AFRL/RIEA. For more information on Palladyne AI and its patented collaborative autonomy software, including SwarmOS, please visit www.palladyneai.com. For more information about AFRL, please visit www.afrl.af.mil. About Palladyne AI Palladyne AI is a U.S.-based technology company developing patented embodied artificial intelligence, collaborative autonomy solutions, advanced avionics, autonomous systems, advanced UAV engineering services, and precision-manufactured components for defense and industrial markets. Palladyne AI delivers secure, American-developed and operated platforms designed to meet the stringent requirements of U.S. government and public-sector customers, including data sovereignty, security, and compliance. Palladyne AI's embodied AI is designed to operate in complex, contested, and high-risk environments, enabling distributed tasking, human-on-the-loop decision-making, degraded-communications resilience, and multi-domain coordination. Its platform-agnostic autonomy stack combines real-time sensor fusion, adaptive AI models, and edge-native orchestration—without vendor lock-in—to support autonomous and collaborative systems across air, ground, maritime, and industrial domains w
ACCESS reveals experimental results on remote robot control and local-like operations using VLA model and IOWN APN.
ACCESS has announced the successful execution of remote robot control experiments utilizing the Vision-Language-Action (VLA) model. Conducted recently, these experiments demonstrated that by leveraging a high-quality communication network, the responsiveness and operational quality of robot control in remote environments can closely match that of local settings for certain tasks. This advancement highlights the potential for improved remote robotic applications, driven by enhanced communication technologies.
ABB Robotics and PSYONIC Use Human-Generated Data to Advance Robotic Dexterity
ABB Robotics has partnered with California-based bionics company PSYONIC to enhance robotic dexterity and grasping capabilities by utilizing human-generated data from prosthetic use. Announced on June 16, 2026, this collaboration aims to address the significant challenge of replicating human-like dexterity in industrial robotics, which is essential for the development of Autonomous Versatile Robotics (AVR™). By integrating the PSYONIC Ability Hand with ABB's GoFa™ collaborative robot, the two companies will explore how real-world manipulation data can train robots to perform delicate tasks that are typically difficult to automate. This initiative is expected to reduce engineering time by up to 30% and improve productivity, flexibility, and workplace safety across various industries, including automotive, aerospace, packaging, logistics, and life sciences. Marc Segura, President of ABB Robotics, emphasized the importance of bridging the gap between human and robotic dexterity to enable robots to learn and interact with their environments more intuitively. Dr. Aadeel Akhtar, Founder and CEO of PSYONIC, highlighted that the collaboration will leverage high-fidelity data on movement and grip force to enhance robotic performance in complex tasks. The GoFa™ robot will provide the precision necessary for industrial applications, ensuring consistent execution of intricate movements, which is crucial for handling fragile or irregular objects. This partnership represents a significant step towards advancing physical AI in robotics, allowing for more effective collaboration between humans and machines.
Zhi Ping Fang launches NeuroVLA, a brain-like embodied intelligence system.
Recently, Zhi Ping Fang unveiled NeuroVLA, an innovative brain-inspired embodied intelligence system, marking a significant advancement in robotics. This system features a unique three-tier architecture that mimics the human brain, comprising the cortex for semantic understanding and task planning, the cerebellum for high-frequency motion coordination and dynamic correction, and the spinal cord for rapid motion execution and safety reflexes. Experimental results indicate that NeuroVLA can reduce robotic motion jitter by over 75% and achieve reflex responses within 20 milliseconds following a collision, all while significantly lowering system power consumption. This breakthrough aims to enhance the efficiency and safety of robotic operations, positioning NeuroVLA at the forefront of intelligent robotics technology.
AGIBOT WORLD CHALLENGE 2026 Advances Embodied AI Competition from Simulation to Real
The AGIBOT WORLD CHALLENGE 2026 took place in Vienna, showcasing a pivotal shift in embodied artificial intelligence as it moved from simulation-based assessments to real-robot testing. The event attracted 526 teams from 27 countries, competing across two distinct tracks: Reasoning to Action and World Model. This competition aimed to address practical deployment requirements by emphasizing the execution of real-world tasks and adaptability of AI systems. The focus on tangible applications is expected to significantly enhance the evaluation framework for embodied AI, marking a notable advancement in the field.
Embodied AI Robotics Artificial Intelligence Technology Competition Benchmarking
NC AI develops welding AI for Hanwha Ocean
NC AI, the artificial intelligence division of gaming company NCSoft, announced on Thursday that it has secured a project with Hanwha Ocean to create AI-driven autonomous welding technology for shipbuilding. This initiative will utilize a vision-based welding model alongside a collaborative robot system, enabling the identification of welding targets and execution of tasks with reduced human involvement. The innovative technology is set to be implemented at Hanwha Ocean's production facilities for both commercial and specialized vessels, aiming to enhance efficiency and precision in the welding process.
All News
The Roboteon Difference
Roboteon is making significant strides in the warehouse automation market with its innovative solutions that emphasize interoperability across various Warehouse Management Systems (WMS) and robotic platforms. The company has developed a comprehensive and functionally-rich platform that encompasses everything from core integration to advanced execution, optimization, and orchestration of tasks. By leveraging artificial intelligence and advanced simulation techniques, Roboteon enhances decision-making and operational efficiency within warehouses. Their deep domain expertise, particularly in integrating robots with existing WMS and automation systems, positions them as a leader in this rapidly evolving sector.
Apellix Spray Painting Drone Goes to Work in the Field for State DOTs, Federal Contractors, and Commercial Coating Firms
A new technological advancement is transforming the way infrastructure, industrial, and government projects are executed, as the Semi-Autonomous Aerial Coating System begins to replace traditional lifts and scaffolding. This innovative system is being deployed across various real-world applications, streamlining the coating process and enhancing efficiency. The shift towards this aerial solution is gaining momentum as organizations seek to improve safety and reduce labor costs associated with conventional methods. By utilizing advanced robotics and automation, the Semi-Autonomous Aerial Coating System allows for precise application of coatings while minimizing the need for extensive manual labor. As this technology continues to evolve, it is expected to play a significant role in modernizing project execution and addressing the challenges faced in construction and maintenance tasks.
Robot.com Launches R-ads, Redefining Out-of-Home Advertising With Its Autonomous Robot Media Network
Robot.com has unveiled R-ads, an innovative advertising platform that transforms autonomous robots into a measurable out-of-home media network. Launched on May 21, 2026, the platform aims to enhance physical advertising by making it as accountable as digital formats. This initiative follows the company's successful execution of over 100 brand activations across more than 20 countries, including high-profile events such as sports leagues and global tech conferences. R-ads operates on the premise that interactive advertising can significantly boost engagement. With over 500 robots currently deployed in various settings, including campuses and city streets, the platform has completed 2.5 million tasks. Judah Longgrear, co-founder of Robot.com, emphasized that R-ads combines the reach of traditional billboards with the interactive capabilities of robots, allowing brands to measure every impression effectively. The platform offers a self-serve advertising suite that includes moving robots, vehicle wraps, and digital screens, enabling campaigns to launch in minutes with real-time analytics. Notably, R-ads recently partnered with the Ad Council for a 15-day activation in Miami, coinciding with National Heatstroke Prevention Day. This campaign utilized the robots' digital displays due to venue restrictions, generating over 147,000 impressions within the first four days. Robot.com’s CEO, Felipe Chavez, highlighted that R-ads exemplifies the multifunctionality of robots, which can deliver, advertise, and collect data, ultimately making robotic technology more accessible and economically viable.
Rapid Stiffness Switching Technology: Peking University Team Develops Biomimetic Octopus Gripper to Solve Underwater Soft Robot's Slow Response Issue
A research team from Peking University, headed by Professor Xie Guangming, has developed a groundbreaking biomimetic octopus gripper capable of transitioning between soft and hard states in less than one second. This advancement enhances the efficiency of underwater robotic operations, enabling more effective and energy-efficient execution of tasks. The team's findings were published in the journal 'Cyborg and Bionic Systems', highlighting the potential applications of this technology in various underwater environments.
Underwater Robotics Soft Robotics Biomimetic Design Robotic Grippers
Major Release: ZhiYuan Research Institute Unveils RoboClaw Embodied Intelligence Operating System
The ZhiYuan Research Institute, in partnership with Xuanji Intelligent and Mianbi Intelligent, has unveiled the RoboClaw operating system, a groundbreaking development aimed at enhancing embodied intelligence in robotics. Launched recently, this innovative system allows robots to autonomously perform tasks by engaging in natural language interactions, effectively bridging the gap between artificial intelligence comprehension and practical execution in the physical world. The initiative reflects a significant advancement in robotics technology, aiming to improve the efficiency and versatility of robots in various applications.
Embodied Intelligence Robotics AI Automation Natural Language Processing
6-Axis Cobot Arms vs. Used/Traditional 6-Axis Robot Arms: Which is Better?
JAKA, a leader in modern manufacturing automation, is advancing efficiency and product quality through the integration of 6-axis robot arm systems and collaborative robot arm solutions. The company emphasizes that the choice between these technologies hinges on specific production requirements, workspace limitations, and safety considerations. The JAKA Zu18 system exemplifies the benefits of contemporary 6-axis robot technology, offering high precision and operational consistency. This system minimizes defects and reduces costs by performing complex tasks, such as intricate spraying and handling delicate components, which traditionally required manual labor. By automating these processes, JAKA not only enhances production rates but also fosters safer and more ergonomic working conditions for operators. Collaborative robot arm systems, like the JAKA Zu18, provide significant advantages in flexible deployment and intelligent control. Their compact design allows operation in confined spaces, and user-friendly interfaces enable operators with minimal robotics experience to engage with the technology. Advanced algorithms enhance task precision and safety, facilitating seamless integration into existing production lines without extensive downtime. Safety remains a crucial factor in the comparison of robotic systems. Unlike traditional robots that necessitate extensive safety barriers, JAKA's collaborative arms can operate alongside humans, thanks to sophisticated sensors and protective features. This capability, combined with integration into existing manufacturing execution systems and automated guided vehicles, allows for adaptable production environments. In summary, JAKA advocates for a thoughtful evaluation of production needs when selecting between 6-axis and collaborative robot solutions. By prioritizing precision, safety, and flexibility, the JAKA Zu18 is positioned as an ideal choice for various manufacturing tasks, ultimately driving down operational costs and enhancing product quality.
Top 5 Reasons to Upgrade Your Old Arm Robot Industrial System to a Cobot
Manufacturers are increasingly recognizing the limitations of traditional industrial robot arm systems, which were originally designed for fixed tasks and isolated work environments. These older systems often lack the flexibility, safety, and user-friendliness required in today’s dynamic production settings. JAKA, a leader in robotics, advocates for upgrading to collaborative robots (cobots) as a solution to these challenges, allowing manufacturers to enhance their operations without overhauling existing workflows. The JAKA Pro5 industrial robot arm exemplifies this evolution, offering improved precision with positioning accuracy of up to 0.2 mm, essential for tasks like welding and assembly. Unlike older systems that may experience performance drift, the Pro5 maintains operational stability through advanced closed-loop control, ensuring consistent production quality. Additionally, its force control capabilities and support for varying payloads enhance process consistency, allowing for reliable execution of tasks even as production demands shift. The Pro5 also streamlines deployment, requiring less complex configuration and shorter commissioning times compared to traditional robots. Its user-friendly interface facilitates communication with welding machines, while built-in safety features ensure secure operations. Furthermore, the system simplifies welding processes through an app that supports various functions, including manual control and debugging modes, which allow operators to verify settings safely. By transitioning to cobots like the JAKA Pro5, manufacturers can achieve greater precision, usability, and adaptability, ultimately fostering a safer and more flexible production environment while maintaining familiar processes.
JAKA Robotics: Driving Innovation as a Leading Robotics Company in APAC
JAKA Robotics is making significant strides in the field of industrial automation, focusing on smart manufacturing solutions across the Asia-Pacific (APAC) region. The company specializes in collaborative robots that work safely alongside human operators, ensuring consistent performance in various applications such as electronics assembly, automotive component handling, and semiconductor processes. With a commitment to precision and stability, JAKA employs advanced control systems and intelligent software, achieving high repeatability levels essential for sensitive tasks like shaft assembly. Their robots are designed with low-noise circuitry and automated quality controls, enabling manufacturers to maintain efficiency and reduce errors during production. The Zu series of JAKA robots highlights the company's emphasis on flexible deployment, featuring compact designs that fit into confined workspaces. User-friendly programming tools allow even those without technical backgrounds to manage complex tasks, while safety features like collision protection and zero-drift calibration enhance human-robot collaboration. JAKA also prioritizes seamless system integration, offering communication interfaces compatible with manufacturing execution systems (MES), automated guided vehicles (AGVs), and vision modules. This adaptability supports more complex automation scenarios and allows non-specialized staff to operate sophisticated processes efficiently. As JAKA Robotics continues to innovate, it remains dedicated to enhancing productivity and operational resilience in the electronics, automotive, and semiconductor sectors, exemplifying how advanced automation solutions can transform modern manufacturing in the APAC region.
JAKA Zu Series vs. JAKA S Series: Which Cobot Suits Your Precision Needs Best?
JAKA, a leader in automation solutions for modern manufacturing, is showcasing its range of collaborative robots designed to enhance precision, flexibility, and safety in production environments. The JAKA Zu Series, particularly the Zu7 model, is tailored for high-precision tasks such as painting and packaging, effectively replacing manual labor while improving production rates and reducing material waste through its accurate control and repeatability. In contrast, the JAKA S Series focuses on smart control and safety, featuring advanced force-sensing and collision protection algorithms that enhance human-robot interaction. This series is designed for easy programming, allowing operators without coding experience to manage complex tasks, making it ideal for environments with limited space or frequently changing processes. Both series facilitate extensive system integration, connecting seamlessly with manufacturing execution systems (MES), automated guided vehicles (AGVs), and vision equipment. This integration supports a flexible manufacturing ecosystem, optimizing productivity and reducing deployment time. As clients evaluate their options, JAKA emphasizes the importance of selecting the right cobot based on specific operational needs. The Zu Series is recommended for high-volume, repetitive applications, while the S Series is better suited for adaptable operations requiring close collaboration with human workers. By understanding the unique strengths of each series, JAKA aims to help clients enhance their production efficiency and meet precision requirements in modern manufacturing.
JAKA S Series vs. JAKA Pro Series: Choosing the Right Industrial Cobot
JAKA, a leader in industrial automation, has introduced two distinct lines of collaborative robots, the S Series and Pro Series, designed to meet varying operational demands on production lines. The S Series focuses on advanced human-robot interaction, featuring cutting-edge force control technology that enhances safety and precision for tasks requiring delicate handling, such as assembly and hand-guided teaching. This series is ideal for environments where direct collaboration with humans is essential. Conversely, the Pro Series is engineered for harsh industrial settings, boasting an IP68 rating for waterproof and dustproof capabilities. This robustness allows it to operate effectively in challenging conditions, such as machining and foundries, where exposure to contaminants is a concern. Both series are equipped with extensive communication interfaces, facilitating integration with manufacturing execution systems and other peripherals. The choice between the two depends on the specific needs of the application—whether the priority is on sensitive interaction or durability in demanding environments. JAKA aims to provide tailored solutions that enhance productivity and reliability in modern manufacturing processes, allowing businesses to select the most suitable robot for their operational context.
Video Friday: Autonomous Robots Learn By Doing in This Factory
In a recent edition of Video Friday, IEEE Spectrum robotics showcased a variety of innovative robotics projects and developments. Notably, scientists at the Toyota Research Institute are collaborating with Toyota Manufacturing to implement autonomous robots on factory floors, enhancing production efficiency. Zipline shared insights into their drone delivery system, detailing the challenges faced and lessons learned in its development. Humanoid introduced KinetIQ, an AI framework designed for the orchestration of humanoid robot fleets, which integrates task allocation and execution through advanced cognitive layers. Meanwhile, researchers at VISTEC unveiled a decentralized adaptive resilient neural control system (DARCON) that enables legged robots to autonomously adapt to limb loss, ensuring mission success despite mechanical failures. NASA's Jet Propulsion Lab presented an animation of the Perseverance rover's drive along the Jezero Crater, captured on December 10, 2025, showcasing the rover's navigation capabilities. Additionally, Unitree's humanoid robot G1 made its mark on the snowfields of Altay, demonstrating its adaptability in extreme conditions. The Norwegian University of Science & Technology introduced a hierarchical 3D scene graph to enhance autonomous agents' understanding of their environments. Other highlights included the HoLoArm quadrotor, which features compliant arms for improved stability, and SkyDreamer, a pioneering vision-based autonomous-drone racing policy. The event also featured demonstrations of dexterous object manipulation by the AI Worker robot and advancements in swarm robotics for architectural applications.
Video-friday Autonomous-robots Humanoid-robots Industrial-robots Robot-ai Perseverance-roverRobotics needs a service framework.
RSF defines a common language for robot service capability, lifecycle operations, certification pathways, and service-provider networks.
