Robotics coverage relating to manipulation.
Before humanoid robots mature, Physical AI is attempting to enter industrial scenarios still reliant on manual operations by deploying on mature robotic arms. Companies like Sanctuary AI and Skild are integrating AI models into robotic arms from different manufacturers to perform complex tasks such as wire insertion.
The article discusses the surge in welding robot orders in the shipbuilding industry, driven by the need for automation and efficiency. It highlights the growth of the market, key players like Fao Robot and Zhongxun Technology, and the technological advancements in AI-driven systems and collaborative robots. The text also addresses challenges in complex welding scenarios and the importance of adaptability and reliability in robotic solutions.
Pudong is accelerating the formation of an embodied intelligence industry ecosystem, driving the deployment of humanoid robots in commercial, public services, and rehabilitation medical scenarios. Companies such as Zhiyuan, Furuiyi, and Mifeng Technology have showcased multiple robot products and data platforms, promoting embodied intelligence technology from laboratories to large-scale applications.
Sharpa has launched three new products: the W02 high-degree-of-freedom tactile hand, the D01 humanoid robot, and the AE01 data glove. These products are designed to enable advanced dexterous manipulation and real-world deployment, with the W02 offering improved tactile sensing and reduced size, the D01 providing high-speed and precise motion control, and the AE01 facilitating high-quality data collection for AI training. The products are positioned as foundational components for integrating large models into physical environments.
The article discusses Sharpa's new robotic hardware, including a dexterous hand, a full-body robot, and a data-gathering glove, emphasizing the importance of integrated sensing and mechanical design for dexterous manipulation. It explores the role of robotic arms, tactile perception, and data acquisition in achieving precise and adaptive robotic operations.
NASA's Dexterous Robotics Team, based at Johnson Space Center, is working on advanced robotic systems to support human spaceflight and exploration. The team focuses on creating highly capable, reliable robots that can perform tasks in extreme environments, with a particular emphasis on humanoid robots and the iMETRO facility for testing and adapting robotic technologies for space applications.
Researchers from North Carolina State University have developed a soft, non-electronic robot that jumps continuously using infrared light. The 'soft ring jumper' uses elastic energy stored through torsion, with a V-shaped aluminum stopper controlling the direction of movement. By adjusting the geometry of the V-stop, the robot can crawl, jump forward, or leap vertically. The design enables autonomous jumping without external resets, opening possibilities for use in environmental navigation and unstructured terrain.
Carnegie Mellon University researchers have developed LAMP, a system that allows multiple robots to work together to move objects through crowded spaces. LAMP combines learned models with search-based planning to enable efficient navigation and coordination, with successful testing in complex environments and a demonstration at the 2026 IROS conference.
This research introduces a novel algorithm for generating grasp keypoints using straight skeletons to identify high-potential regions for stable grasping. A fully automated pipeline enables large-scale annotation without human intervention, and a new architecture, SkelGG-CNN, is introduced for 4-DoF grasping. The method is extended to 6-DoF grasping using a generalized quasi-3D skeletonization algorithm, achieving high accuracy on multiple benchmarks and real-world experiments. The framework enables efficient training of lightweight models for real-world deployment.
Aletta, the first autonomous blood-draw device approved for use in the U.S., uses imaging and robotics to perform blood draws with high success rates, even in challenging cases. While it addresses staffing shortages in clinical labs, concerns about skin tone bias and sample quality persist, requiring further validation.
Researchers at the University of Queensland have developed 'Paraborgs'—cyborg cockroaches equipped with needles and cameras to assist in search-and-rescue missions. These insects, grafted with electronic interfaces, can navigate rubble, deliver drugs, and operate in teams. The study addresses challenges in autonomy, communication, and human interaction with the technology.
Skild AI introduces S1, a robotic foundation model that leverages in-context learning to execute complex, long-horizon tasks without post-training. This marks a significant shift from traditional fine-tuning approaches, enabling rapid deployment and reducing data requirements. The model demonstrates strong performance on unseen tasks, including plant potting, pancake cooking, and kit assembly, and shows robustness to perturbations and common-sense reasoning.
This article explains how tactile sensing allows robots to detect and correct off-center grasps by analyzing force distribution across fingertips. It describes a demonstration using a 2F-85 gripper with TSF-85 tactile sensors, which repositions and regrasps objects to achieve a balanced grip. The method improves grasp stability and reduces the need for reorientation or regrasping.
Engineers at MIT and EPFL have developed a flapping-wing robot, or FAAV, that can swim underwater and then flap into the air, mimicking the movement of diving birds. The robot, weighing less than 300 grams, can transition between water and air using optimized wing size, flapping frequency, and tail angle. The study, published in Science, could enable new aerial-aquatic drones for oceanographic research and environmental monitoring.
Figure's latest generation robot, F.03, has arrived at BMW Group Plant Spartanburg, demonstrating advanced dynamic whole-body control and sequencing use cases. Powered by Helix 02, the robot performs complex material manipulation tasks that require both precision and force, marking progress in embodied AI for manufacturing.
Figure AI demonstrates Helix-02 humanoids performing a bedroom reset in under two minutes using a single learned Vision-Language-Action policy. The robots coordinate without shared planners or message passing, inferring each other's intent from motion alone. Tasks include opening doors, pushing furniture, draping clothing, and making a bed, showcasing integration of locomotion, dexterity, and sensing.
Dexterity introduces Foresight, a world model that enables robots to reason about the physical world, predict outcomes, and act confidently in real environments. Trained on over 100 million autonomous actions in production, Foresight supports predictive branching, pragmatic decision-making, capability-aware orchestration, and predictive pipelining. It is designed to be interpretable, safe, and fast, with applications in logistics and industrial automation.
Physical AI refers to AI systems that enable robots to perform physical tasks in the real world. The article explains the immense difficulty of this endeavor, highlighting the variability of the physical environment, the need for multiple AI capabilities to work together, and the importance of safety and reliability. Dexterity's compositional approach to AI architecture is presented as a solution that enables production-scale Physical AI across multiple industries.
Dexterity and ASRock Rack have partnered to develop custom AI supercomputing systems for Dexterity’s Mech industrial superhumanoid robot, enabling real-time edge intelligence for warehouse automation tasks. These edge AI servers are optimized for computational demands, supporting complex reasoning and manipulation at the edge without cloud dependency.
Dexterity's Mech industrial superhumanoid has been approved for operational validation at Sagawa Express' X Frontier® relay center in Tokyo, Japan. This marks the start of onsite use beginning May 2025, following successful testing. Mech's deployment aims to address Japan's logistics crisis by automating truck loading tasks with Physical AI, which meets and exceeds performance benchmarks in speed, quality, and space utilization.