Technical Directions
Robotics coverage relating to sim-to-real.
292 selected reportsLast 30 days 253 recordsAll coverage 1,239 records
NVIDIA has released a reinforcement learning policy called SO-101 Reach, which enables the SO-101 robot arm to reach randomly sampled 3D goal positions using delta joint actions. The policy is trained entirely in simulation and is intended for sim-to-real transfer and research purposes. It is available as an ONNX model and includes an Isaac Lab I/O descriptor.
Researchers at MIT have developed FloatForm, a system of robotic tiles that can autonomously assemble into floating structures. Inspired by ant colony behavior, the system uses magnetic latches, sensors, and thrusters to reconfigure dynamically, addressing challenges like hydrodynamics and slipstreams. Potential applications include emergency response, public space, and infrastructure on water.
The article explores how the DARPA Grand Challenge and Robotics Challenge offer lessons for the development of physical AI. While autonomous driving has benefited from structured environments and abundant data, physical AI must address challenges like diverse physical interactions, limited real-world data, and the need for system reliability in industrial settings. The article emphasizes the importance of transitioning from demonstrations to reliable, economically viable production systems.
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.
NEURA Robotics presents a seven-stage pipeline for training robots, from defining use cases and capturing real-world data to model training, optimization, and deployment. The process emphasizes physical AI, multimodal data collection, and iterative refinement to ensure reliable, scalable robotic capabilities.
FedEx and Dexterity have expanded their collaboration to deploy Dexterity's Foresight world model and Mech trailer loading systems at the FedEx Hagerstown Hub. This marks a significant scaling of physical AI in logistics, enabling production at a larger operational scale with a focus on safety, consistency, and performance.
The US is banning the sale of foreign-made robots, including those from China, due to national security concerns. The FCC cites risks such as vulnerabilities in networked capabilities and potential exploitation of remote connectivity. The ban also extends to inverters with Wi-Fi, cellular, or Bluetooth control. Existing products already sold are unaffected, but companies can apply for approval to continue selling in the US.
LeRobot has converted the Hy-Embodied-0.5-VLA-RoboTwin model, achieving full forward parity with the original policy. The model supports 16D dual-arm pose/gripper actions and passes all validation checks, though it cannot handle more complex action spaces.
LeRobot has released a converted version of the Hy-Embodied-0.5-VLA-UMI model, ensuring full tensor compatibility and validation. The model supports a 16D dual-arm embodiment but cannot express 12D mobile-manipulator actions, indicating limitations in the current representation.
NVIDIA introduces Cosmos-H-Dreams, a real-time, action-conditioned generative simulator for surgical robotics. It distills the capabilities of Cosmos-H-Surgical-Simulator into a causal, few-step student model, enabling interactive environments for policy evaluation and synthetic data generation. The system runs on a single NVIDIA RTX PRO 6000 GPU and supports integration with surgical platforms like Versius.
NVIDIA has released an open-source, GPU-accelerated Medical Physics Simulation framework as part of its Isaac for Healthcare platform. This tool enables developers to model anatomy-device interactions, generate complex scenarios, and train robot policies in simulation, significantly reducing development time and improving regulatory readiness.
MIT's JARVIS Challenge explored whether AI can transform complex engineering tasks like building jet engines. Students used AI tools to design, fabricate, and test small gas turbine engines, but found that human expertise and judgment remained critical. The challenge revealed that while AI can speed up design and testing, physical manufacturing and engineering experience are still essential. The study emphasizes the need for a balance between AI assistance and human oversight in engineering workflows.
MIT researchers have developed SceneSmith, a system that uses AI agents and vision-language models to generate highly realistic and detailed virtual environments for robot training. These environments allow robots to practice tasks in simulated settings, reducing the need for physical testing. The system can create complex 3D scenes with a high density of objects, ensuring physical accuracy and realism, which helps improve robot performance in real-world scenarios.
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.
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.
NVIDIA and Hugging Face are releasing the NVIDIA Isaac GR00T 1.7 model and Isaac Teleop framework into LeRobot, an open-source robotics library, to provide developers with shared tools for training, evaluating, and deploying robot foundation models. NVIDIA Cosmos 3, a frontier world model for physical AI, is also set to be integrated soon.
LeRobot v0.6.0 introduces world model policies that learn to imagine the future, new VLAs, and a reward models API. It also includes six new simulation benchmarks, improved dataset handling with depth support and language annotations, and deployment tools like lerobot-rollout.
Apptronik announced the opening of its expanded Robot Park in Austin, Texas, and the launch of Apollo 2, a modular humanoid robot platform. The facility and robot are part of a global network, enabling real-world data collection and training for embodied AI models. Apollo 2 supports both bipedal and wheeled configurations, with data used to refine Gemini Robotics models for commercial deployment.
Firefly Aerospace's Blue Ghost Mission 2 will deploy NVIDIA Jetson for on-orbit AI processing, enabling real-time lunar data analysis and reducing latency. This marks the first use of Jetson in lunar orbit, supporting scientific research and future lunar exploration.
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.
Article dates follow your selected timezone. Briefing editions use Hong Kong time (UTC+8).