Reading Modes
Papers & Evaluations with links to supporting source material.
206 selected reportsLast 30 days 182 recordsAll coverage 1,487 records
Sanctuary AI has demonstrated zero-shot in-hand manipulation using their proprietary hydraulic hand, autonomously reorienting a lettered cube ten times without dropping it. This showcases successful simulation-to-reality transfer, overcoming challenges in simulating dexterous hands and highlighting the company's advanced robotic hardware and policy training capabilities.
Universal Robots and Scale AI have unveiled the UR AI Trainer, an imitation learning system designed to accelerate AI model training by capturing high-fidelity, synchronized robot and vision data. This system enables direct lab-to-factory AI training using UR robots with Direct Torque Control and force feedback, addressing challenges in data capture and hardware fragmentation. The collaboration also includes a large-scale industrial dataset and showcases real-world robotic foundation model performance.
NASA showcased its robotic technologies and future space exploration vision at the 2025 FIRST Robotics World Championship, engaging over 55,000 students and 75,000 parents and mentors. The event featured interactive exhibits, including lunar rovers, robotic arms, and the Mars Perseverance Rover, with NASA centers highlighting innovations like automated assembly systems and surface manipulation technologies.
Sanctuary AI has demonstrated advanced manipulation skills using reinforcement learning to train dexterous policies for their high-performance hydraulic hands. The system successfully reorients objects in the real world under gravity and with added weight, showcasing the effectiveness of their approach. This represents the first successful transfer of such policies to five-fingered hands, highlighting the company's proprietary technology and hardware capabilities.
Sanctuary AI has demonstrated industry-leading sim-to-real transfer of dexterous manipulation policies using their high-degree-of-freedom hydraulic hands. The company leverages NVIDIA Isaac Lab, an open-source framework for robot learning, to simulate dexterity-focused training environments. This approach enables efficient training of autonomous robots, with the ability to train thousands of hands simultaneously and accelerate the learning process through online reinforcement learning.
NIST and IEEE have developed a standard for robotics ontologies, designed to improve robot programming, communication, and task execution. The standard, known as IEEE P1872, provides a formal framework for representing robot knowledge and enables efficient information exchange. It is a foundational step toward creating a shared understanding among robots and humans.
Article dates follow your selected timezone. Briefing editions use Hong Kong time (UTC+8).