Technical Directions
Robotics coverage relating to embodied perception.
190 selected reportsLast 30 days 148 recordsAll coverage 519 records
Michigan Robotics is participating in the 2026 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) in Pittsburgh, presenting research on topics such as robotic leg prostheses, tensegrity robotics, and soft robot control. The event includes workshops, papers, and opportunities for networking and exploring Pittsburgh.
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.
Cornell researchers and collaborators have developed a snail-inspired robot called Mollusca to filter microplastics from water. Teachers from New York State participated in workshops to build and test the robots, integrating the project into their classrooms. The initiative aims to educate students on environmental issues and robotics through hands-on learning and real-world applications.
Google DeepMind has released Gemini Robotics 2, a new model that enables robots to perform complex tasks with whole-body intelligence, advanced dexterity, and multi-robot collaboration. The model supports real-time reasoning, on-device adaptation, and safety features for human-robot interaction.
This article explores the legacy and impact of the DARPA Robotics Challenge, which aimed to develop robots capable of performing complex tasks in disaster scenarios. It discusses key lessons learned, including the importance of shared autonomy, mechanical design, and reliability in robotic systems. The piece also examines the long-term impact of the DRC on industry and research, emphasizing how it shaped the development of physical AI and its current applications.
1X announces the release of its 25 Degree of Freedom (DOF), tendon-driven hands for the NEO humanoid platform, achieving near human-level dexterity, strength, safety, and reliability. These hands are designed to remove the hardware ceiling on what humanoid robots can do, enabling precise, adaptable, and gentle interaction with the real world.
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.
University of Michigan researchers have developed SPPIRO, an untethered soft robot inspired by earthworms, capable of navigating complex pipelines with varying diameters and sharp turns. Equipped with Kresling origami actuators, onboard sensors, and a self-mapping system, SPPIRO can inspect pipelines at speeds up to 9.5 mm/s vertically and 6.5 mm/s horizontally, outperforming existing untethered soft in-pipe robots. The robot uses liquid-metal-based sensors and an onboard camera to detect defects in enclosed pipes, though challenges remain in maintaining stability and accuracy in certain conditions.
Google DeepMind has released Gemini Robotics-ER 1.6, an upgraded model that enhances embodied reasoning for robotics. This model improves spatial reasoning, multi-view understanding, and safety compliance, enabling robots to perform complex tasks like instrument reading. It is available via the Gemini API and Google AI Studio, with examples provided in a developer Colab.
NVIDIA has released SOMA-X v0.3.0, a unified framework for parametric human body models that decouples identity representation from pose parameterization. The release supports six identity backends including SOMA-shape, SMPL, and SMPL-X, and introduces native hand layers with interoperability for user-supplied MANO models. Designed for commercial use under the Apache 2.0 license, the model enables real-time simulation pipelines on NVIDIA GPUs with throughput exceeding 7,000 meshes per second on an A100. The core pipeline relies on closed-form analytical modules rather than learned neural networks, though an optional shallow MLP can be enabled for surface correction.
Article dates follow your selected timezone. Briefing editions use Hong Kong time (UTC+8).