Articles tagged with "robotic-arms"
Physical Intelligence, Stripe veteran Lachy Groom’s latest bet, is building Silicon Valley’s buzziest robot brains
Physical Intelligence, co-founded by UC Berkeley associate professor Sergey Levine and led by Silicon Valley veteran Lachy Groom, is developing advanced robotic "brains" designed to enable robots to perform everyday tasks with human-like adaptability. Operating out of a modest San Francisco facility, the company uses off-the-shelf robotic arms priced around $3,500, emphasizing that sophisticated intelligence can compensate for relatively inexpensive hardware. Their approach involves training general-purpose robotic foundation models through continuous data collection from various environments, including warehouses, homes, and test kitchens. These models are then tested on tasks such as folding clothes or peeling vegetables, with the goal of enabling robots to generalize learned skills to new, unseen objects and challenges. The company’s work is likened to creating a "ChatGPT for robots," where the AI learns from diverse real-world interactions to improve its capabilities. Physical Intelligence’s experimental setup includes multiple robotic arms attempting mundane tasks, illustrating the iterative process of refining the models. Lachy Groom, a young entrepreneur with a background
roboticsartificial-intelligencerobotic-armsautomationmachine-learningrobotics-researchrobot-trainingRobot Talk Episode 142 – Collaborative robot arms, with Mark Gray - Robohub
The article summarizes Episode 142 of the Robot Talk podcast, featuring a conversation between host Claire and Mark Gray from Universal Robots. Mark Gray, with 30 years of experience in automation and robotics, discusses Universal Robots' lightweight collaborative robot arms (cobots) designed to work safely alongside humans. As the UK country manager and the company's first UK employee, Gray has led projects with prominent research institutions such as the Advanced Manufacturing Research Centre (AMRC), The Manufacturing Technology Centre (MTC), the National Robotarium, and Bristol Robotics Lab, highlighting the integration of cobots in advanced manufacturing and research environments. Robot Talk is a weekly podcast exploring robotics, AI, and autonomous machines, with recent episodes covering diverse topics such as human interaction with robot swarms, robotic agility inspired by animals, advanced robot hearing, and AI-powered robotic dogs for emergency response. The podcast serves as a platform to showcase cutting-edge developments in robotics and their practical applications across various fields.
roboticscollaborative-robotscobotsautomationartificial-intelligencerobotic-armshuman-robot-interactionGenerative-AI robot unloads trailers at 1,500 boxes per hour
The Pickle Robot Company has developed AI-powered, one-armed robots that autonomously unload trailers at rates of up to 1,500 boxes per hour, lifting boxes weighing up to 50 pounds. These robots combine generative AI, machine learning, sensors, cameras, and machine vision to quickly adapt to new warehouse environments and improve performance over time. Their hardware includes industrial robotic arms typically used in car manufacturing. Early deployments with companies like UPS, Ryobi Tools, and Yusen Logistics have demonstrated that these robots reduce the physical strain on human workers, allowing humans to focus on complex problem-solving tasks rather than repetitive, injury-prone labor. Founded by MIT alumni AJ Meyer, Ariana Eisenstein, and Dan Paluska, Pickle Robot Company pivoted from sorting robots to unloading systems after initial funding challenges. A proof-of-concept video showcasing a robot unloading trucks sparked customer interest and investor support. The robots use a KUKA arm mounted on a mobile base that autonomously drives into trailers
roboticsgenerative-AIwarehouse-automationlogistics-robotsmachine-learningrobotic-armsindustrial-automationVideo: Baseball-playing robot swings, hits and catches with pinpoint accuracy
Researchers at the RAI Institute in Cambridge, Massachusetts, have developed two robots capable of playing baseball with human-like speed and precision. These robots can throw, catch, and hit baseballs, demonstrating advanced reflexes and control. The system uses soft joints and adaptive control to absorb impact forces, allowing smooth and safe interactions with the ball. Sensors track the ball’s trajectory, while prediction algorithms enable the robots to position their gloves accurately and respond almost instantly to throws. The robots can throw at speeds up to 70 mph and catch or hit balls thrown at speeds up to 41 mph and 30 mph respectively, from distances around 7 meters. The robots’ arms are designed with lightweight materials like carbon fiber rods and flexible joints, combined with electric actuators for quick, smooth movements. Their low center of gravity enhances stability during swings and catches. Using ordinary computer hardware, the system integrates live camera feeds with motion models to guide precise movements. The software continuously adjusts for the ball’s unpredictable spins and curves
roboticsrobotsrobotic-armsartificial-intelligenceautomationsensorselectric-actuatorsPartner Robotics picks up funding to deploy more construction robots - The Robot Report
Partner Robotics, a Dongguan-based company founded in 2023 by Kecheng Wang, recently closed a Series A funding round in the eight-figure RMB range (around $1.4 million USD), contributing to a total of approximately RMB 100 million ($14 million) raised since inception. The new capital will support the development and commercialization of embodied intelligence technologies for construction, expansion into overseas markets including Europe, North America, and the Middle East, and strengthening its supply chain for improved quality and cost efficiency. The funding round was led by China Growth Capital with participation from existing investors Cowin Capital and Redpoint China Ventures. The company currently focuses on two main robotic products: the Floor Tile Paving Robot P900 and the Intelligent Scribing Robot L3000. The P900 automates indoor tile laying with precision controls, reportedly working five to six times faster than human labor while reducing defects like hollow spots. The L3000 supports multiple construction phases with accuracy to within 2 mm and efficiency gains
roboticsconstruction-robotsautomationembodied-intelligencemulti-sensor-perceptioncloud-based-simulationrobotic-armsChina builds high-precision robot arms for nuclear reactor upkeep
Researchers at China’s Hefei Institutes of Physical Science have developed an AI-driven robotic system capable of performing highly precise maintenance tasks inside fusion reactors, achieving accuracy within 0.1 mm. This system addresses the challenging “peg-in-hole” assembly task critical for replacing reactor components, which traditionally requires slow, manual intervention. Utilizing deep reinforcement learning (DRL), the robot mimics human hand-eye coordination by integrating data from a 2D camera and force/torque sensors, avoiding the unreliable 3D sensors that struggle in the reactor’s high-radiation, reflective environment. To support this AI, the team engineered robust hardware, including a novel robotic joint with an ultra-high reduction ratio (13,806:1) delivering 139 kNm torque, enabling precise manipulation of large reactor components. Additionally, they developed TCIPS, a Transformer-based AI perception model that processes 3D point cloud data to segment the reactor interior into basic geometric shapes, enhancing navigation and obstacle avoidance. These innovations collectively
roboticsAIfusion-reactor-maintenancedeep-reinforcement-learningrobotic-armsautomationnuclear-energyReframe Systems wants to use robotic microfactories to change how we build homes - The Robot Report
Reframe Systems, founded in 2022 by Vikas Enti and Aaron Small, aims to address the U.S. housing shortage and improve home resilience through robotic microfactories. The company leverages off-the-shelf robotic arms combined with proprietary vision, motion planning, and fixturing technologies to automate the framing and assembly of homes away from traditional construction sites. This microfactory approach allows Reframe to consolidate operations, build components like walls and ceilings out of the conventional sequence, and target automating 60% to 80% of home construction. The startup recently raised $20 million to scale its operations, with a long-term goal of building one million homes by 2045. Enti highlights that the construction industry's fragmentation—requiring about 25 specialized subcontractors per home—has hindered widespread automation adoption, as most firms operate as small businesses lacking resources for R&D or technology investment. By shifting construction to controlled microfactories, Reframe aims to overcome these challenges and improve
roboticsmicrofactoriesconstruction-automationrobotic-armshome-building-technologysustainable-constructionhousing-innovationChinese team develops drones that exchange tools while flying
Researchers at Westlake University in China have developed the FlyingToolbox system, enabling multi-rotor drones to exchange tools mid-flight with unprecedented precision. This breakthrough overcomes the significant challenge of downwash—the turbulent airflow generated by the propellers of an upper drone that destabilizes a drone flying below it. The system pairs a lower “toolbox drone” carrying multiple tools with an upper “manipulator drone” equipped with a robotic arm that picks up and returns tools during flight. Using real-time airflow prediction via a neural network, magnetic docking mechanisms, and QR code-based visual tracking, the drones achieve sub-centimeter docking accuracy, maintaining stability even in strong turbulence. In experiments, the FlyingToolbox successfully completed 20 consecutive dockings with less than one centimeter of error, a marked improvement over previous aerial docking systems that typically had accuracies of six to eight centimeters. The magnetic docking system, combined with elastic cords, allows for automatic locking and error absorption during alignment. This technology enables drones to
robotdronesaerial-roboticsrobotic-armsdrone-technologymulti-rotor-dronesautonomous-systemsHow satellite refueling in orbit could cut space junk and extend missions
The article discusses the emerging field of satellite refueling in orbit (OOR), highlighting its potential to extend satellite missions, reduce space debris, and promote sustainable spaceflight. While the concept of OOR has existed since the Space Age, recent advances by NASA and commercial companies like Arkisys and Orbit Fab are bringing it closer to reality. Refueling satellites in orbit could significantly extend their operational lifespans, thereby reducing the accumulation of space junk—a growing threat in Low Earth Orbit (LEO) known as the Kessler Effect. Unlike current debris removal efforts that address symptoms, OOR tackles the root cause by shifting satellites away from single-use technology. The article also outlines the technical challenges of refueling satellites, which were not originally designed for in-orbit servicing. Issues include the complexity of robotic grappling, precise navigation, and the lack of visual markers like QR codes on most satellites to aid servicing spacecraft. Despite these hurdles, NASA’s Robotic Refueling Mission (RRM) between
robotspace-technologysatellite-refuelingorbital-debrisspace-sustainabilityrobotic-armsspace-navigationChina tests world-class robot arms for ‘artificial sun’ project
China has successfully tested a state-of-the-art robotic arm system designed for maintaining future fusion reactors, advancing its ambitious “artificial sun” project. The system, developed for the Comprehensive Research Facility for Fusion Technology (CRAFT) under the Chinese Academy of Sciences, features three robotic arms, including a massive manipulator capable of lifting 60 tons—equivalent to the weight of 10 African elephants—with millimeter-level precision. This remote-handling platform is engineered to operate in the extreme conditions inside fusion reactors, such as high heat, intense radiation, and strong magnetic fields, where human intervention is impossible. During testing, the main arm achieved vertical lifting accuracy of 3 to 4 millimeters, while the smaller arms demonstrated positional accuracy within ±0.01 millimeters, making it the most advanced system of its kind in the fusion field. The robotic system addresses critical maintenance challenges posed by the harsh environment inside fusion reactors, where components like cladding and divertors endure constant stress from plasma and radiation
robotfusion-energyrobotic-armsnuclear-fusionprecision-roboticsremote-handlingenergy-technologyUS Navy servicing robot spacecraft endures punishing space trial
The U.S. Naval Research Laboratory (NRL) has successfully completed rigorous space-readiness testing of its Robotic Servicing of Geosynchronous Satellites (RSGS) payload integrated with Northrop Grumman’s Mission Robotic Vehicle (MRV) spacecraft bus. This milestone, achieved after over two decades of development and supported by DARPA funding, involved subjecting the system to extreme thermal vacuum conditions that simulate the harsh environment of space. The RSGS-MRV system is designed to inspect, repair, upgrade, and perform orbital adjustments on satellites operating in geosynchronous orbit—approximately 22,000 miles above Earth—where traditional satellite servicing is currently impossible. Equipped with two robotic arms, advanced avionics, and a suite of cameras, sensors, and infrared imaging tools, the system can safely approach client satellites and execute complex tasks such as capture, inspection, and hardware upgrades. This technology aims to extend satellite lifespans, reduce costs, and enable new innovations by
roboticssatellite-servicingspace-roboticsUS-Navy-technologygeosynchronous-orbitrobotic-armsspace-technology'Self-fixturing’ friction stir welding could soon enter into manufacturing
A breakthrough at the U.S. Department of Energy’s Pacific Northwest National Laboratory (PNNL) promises to expand the use of friction stir welding (FSW) in manufacturing, particularly on automotive assembly lines. FSW is an energy-efficient welding technique that uses a spinning tool to generate heat and deformation, joining metals without rivets or adhesives. However, its adoption has been limited because the process exerts tremendous force—up to 5,000 pounds—that traditionally requires a rigid anvil underneath the materials to constrain and ensure precise welds. This constraint has made it difficult to apply FSW broadly on assembly lines. PNNL researchers have developed a novel “self-fixturing” friction stir welding system that integrates both the spinning tool and a miniature backing plate into a robotic arm attachment. This innovation eliminates the need for a separate rigid anvil, allowing the welding tool to be more maneuverable and adaptable to complex parts. The team is also incorporating a hydraulic system that captures the forces generated during welding,
friction-stir-weldingadvanced-manufacturingrobotic-armsmaterials-scienceenergy-efficiencyautomotive-manufacturingself-fixturing-technologyDog crate-sized robot factory trains itself by watching human demos
MicroFactory, a San Francisco-based startup founded in 2024, has developed a compact robotic system roughly the size of a dog crate that can perform a wide range of manual tasks typically done by human hands. The system features two robotic arms capable of precise operations such as circuit board assembly, soldering, cable routing, and even delicate actions like threading a needle. It is designed to automate repetitive manual labor and can assemble real products efficiently, with the company claiming it is more effective than humanoid robots due to its simpler, non-humanoid design optimized for both hardware and AI. The robotic system can be trained through AI or by human demonstration using an external robotic arm to physically guide the in-box arms through tasks. This teaching method enables the robot to replicate complex motions accurately and learn new tasks quickly. MicroFactory has also developed a user interface that breaks down tasks into smaller steps to facilitate training and operation. Since launching their prototype within five months, the company has received hundreds of preorders from customers
roboticsrobotic-armsAI-trainingautomationelectronics-assemblyMicroFactorygeneral-purpose-robotsThis $30M startup built a dog crate-sized robot factory that learns by watching humans
San Francisco-based startup MicroFactory has developed a compact, dog crate-sized robotic manufacturing system designed for precision tasks such as circuit board assembly, soldering, and cable routing. Unlike traditional humanoid or large-scale factory robots, MicroFactory’s enclosed workstation features two robotic arms that can be trained through direct human demonstration as well as AI, enabling faster and more intuitive programming for complex manufacturing sequences. Co-founder and CEO Igor Kulakov emphasized that this approach simplifies both hardware and AI development while allowing users to observe the manufacturing process in real time. Founded in 2024 by Kulakov and Viktor Petrenko, who previously ran a manufacturing business, MicroFactory built its prototype within five months and has since received hundreds of preorders for diverse applications, including electronics assembly and even food processing. The company recently raised $1.5 million in a pre-seed funding round, valuing it at $30 million post-money, with investors including executives from Hugging Face and Naval Ravikant. MicroFactory plans to
roboticsmanufacturing-automationAI-roboticsrobotic-armstabletop-robot-factoryhuman-robot-interactionprecision-manufacturingRoboBallet makes robotic arms dance in sync on factory floors
RoboBallet is a new AI system developed by a team from UCL, Google DeepMind, and Intrinsic that choreographs the movements of multiple robotic arms on factory floors, significantly improving efficiency and scalability in manufacturing. Traditional robotic coordination requires extensive manual programming to avoid collisions and complete tasks, a process that is time-consuming and prone to errors. RoboBallet overcomes these challenges by using reinforcement learning combined with graph neural networks, enabling it to plan coordinated movements for up to eight robotic arms performing 40 tasks in seconds, even in previously unseen layouts. This approach treats obstacles and tasks as points in a network, allowing rapid and adaptable planning that outperforms existing methods by generating plans hundreds of times faster than real-time. The system’s scalability is a major breakthrough, as it learns general coordination rules rather than memorizing specific scenarios, making it capable of handling complex, dynamic environments where factory layouts or robot configurations change frequently. RoboBallet’s ability to instantly generate high-quality plans could prevent costly
roboticsindustrial-automationAIrobotic-armsmanufacturing-technologyreinforcement-learningfactory-efficiencyRealMan launches robotics data training center in Beijing - The Robot Report
RealMan Robotics, a Beijing-based developer of robotic arms and mobile manipulators, has launched a new robotics data training center in Beijing. The 3,000-square-meter facility integrates core technology R&D, scenario-based application testing, operator training, and ecosystem collaboration. It features 108 diverse robots—including dual-arm mobile manipulators, wheeled semi-humanoids, drone-arms, and quadrupeds—deployed across ten real-world environments such as eldercare, rehabilitation, automotive assembly, and smart catering. These scenarios enable large-scale multimodal data generation, producing over one million high-quality data points annually to train advanced AI models via the newly unveiled RealBOT Embodied Intelligence Open Platform. The center aims to address three key challenges in robotics: lack of cross-scenario data generalization, gaps between simulation and real-world conditions, and the absence of standardized data formats and efficient closed-loop iteration. By creating a full-stack data pipeline from collection to deployment, RealMan seeks to accelerate commercialization of
roboticsrobotic-armsAI-traininghumanoid-robotsdata-acquisitionmobile-manipulatorsrobotics-R&DStäubli Robotics rises to enable baking inspection and handling - The Robot Report
Stäubli Robotics is advancing automation in the baking industry by offering specialized robotic systems designed for food-grade handling, inspection, and packaging. Addressing challenges such as labor shortages, market pressures, and evolving consumer demands for variety and healthier options, Stäubli provides four- and six-axis robots, automated guided vehicles (AGVs), autonomous mobile robots (AMRs), and collaborative robots. These systems enhance production efficiency, throughput, and safety while maintaining flexibility and hygiene standards critical in food environments. Their HE series robots are specifically engineered to withstand rigorous washdown procedures, ensuring operational efficiency and compliance with food safety regulations. At the upcoming International Baking Industry Exposition (IBIE), Stäubli will showcase its TS2-60 HE robot integrated with Oxipital AI’s VX2 vision-guided system. This demonstration highlights a fully automated process where the robot bulk-picks bread products, uses AI-powered inspection to detect defects, sorts acceptable items for packaging, and diverts defective ones for further handling. The
roboticsindustrial-robotsfood-automationbaking-industryrobotic-armsautomated-guided-vehiclescollaborative-robotsChinese firm to deploy 100 robots in car manufacturing factories
Chinese robotics start-up AgiBot (Zhiyuan Robotics) has secured a significant order to deploy nearly 100 of its A2-W dual-armed humanoid robots at Fulin Precision’s automotive parts factories. These robots, standing 5 feet 9 inches tall and weighing 121 pounds, are designed with ergonomic, human-like motion and advanced sensors to handle repetitive and physically demanding tasks such as box-moving and raw material delivery. The deployment aims to enhance production efficiency by enabling the robots to support the manufacturing of 500 units per shift while performing close to 10,000 box-moving actions simultaneously, thereby freeing human workers to focus on higher-value operations and reducing physical strain. This move reflects the growing momentum of industrial robotics adoption in China, where installations rose by 54 percent in 2024 to around 290,000 units, representing over half of global installations. The industrial robotics market in China is projected to reach USD 16.5 billion by 2033, growing at a
robotsindustrial-automationmanufacturing-robotshumanoid-robotsfactory-automationrobotics-in-automotiverobotic-armsPrismaX launches teleop platform for robotic arms - The Robot Report
PrismaX, a San Francisco-based startup with $11 million in funding, has launched a teleoperation platform for robotic arms aimed at bridging the gap between robotics and mainstream adoption. The platform allows users to remotely operate robotic arms and serves as a foundational step toward a future where humans and robots collaborate to enhance human capabilities. PrismaX’s co-founder and CEO Bayley Wang emphasized that this tele-op system is a proof of concept for a labor market where humans and robots work hand in hand, with the company focusing initially on teleoperations and visual data collection to train AI models. PrismaX has outlined a roadmap aligned with the robotics industry's evolution: in the short term, teleoperators will gather data and gain experience; in the mid-term, operators will manage fleets of robots performing real tasks; and in the long term, robots will achieve high autonomy powered by foundational AI models. The company envisions a self-reinforcing "data flywheel" where increased robot operation generates valuable datasets that improve AI,
roboticsteleoperationrobotic-armsAIautomationteleop-platformrobotics-industrymaxon Group acquires minority stake in Synapticon
The maxon Group, a Swiss leader in high-precision drive systems, has acquired a minority stake in Germany-based Synapticon, a provider of intelligent motion control solutions with integrated safety functions. While financial details were not disclosed, the collaboration aims to jointly develop high-performance drive solutions that integrate functional safety by combining components from both companies. This partnership allows maxon to expand its expertise in highly integrated drive systems and offer more powerful and safer solutions to customers, particularly in fast-growing sectors such as robotics, medical technology, intralogistics, and industrial automation. Synapticon’s “software-first” approach focuses on compact, intelligent motion control technologies used in robotic arms, autonomous transport, and complex industrial machinery. The collaboration enables efficient development of existing products and alignment with new applications, addressing increasing functional safety requirements in high-tech industries. Both companies view the partnership as strategically important, with Synapticon’s CEO emphasizing the significant market opportunities and potential to advance motion control technology. Additionally, Synapticon’s CEO
roboticsmotion-controldrive-systemsfunctional-safetyindustrial-automationrobotic-armsautonomous-transport-systems