robotics

A tweet announces that University of Michigan's ROB 501, a 26-lecture series covering the mathematical foundations of robotics, is available for free on YouTube with open-source materials on GitHub.

Kyber Labs demos a robot hand that uses only the force needed for tasks, allowing safe human interruption without injury.

Agility Robotics, maker of the humanoid robot Digit, plans to go public via a SPAC merger with Churchill Capital Corp XI in a $2.5 billion deal, raising over $620 million to scale production of its next-generation robot.

Introducing HIW-500, the largest open-source humanoid teleoperation dataset collected in real homes, with over 500 hours and 23K+ episodes across 12 homes in Southeast Asia.

Honda has developed a precision robotic hand capable of driving tiny screws, showcasing advanced dexterity for manufacturing applications.

MIT researchers have developed a new system-on-a-chip that enables tiny robots to create detailed 3D maps of their environments in real-time using only about 6 milliwatts of power, potentially enabling long-duration autonomous navigation in complex spaces.

InSight presents a framework for autonomous skill acquisition in vision-language-action (VLA) models by enabling steerability at the primitive-action level and using a VLM-guided data flywheel to generate demonstrations, achieving manipulation tasks like block flipping and pouring without human demonstrations.

The paper presents World Value Model (WVM), a generalist robotic value model that combines world models with value estimation to accurately assess task progression and improve robotic policy learning from mixed-quality data, achieving state-of-the-art results on standard benchmarks and a new suboptimal data benchmark.

General Motors installed 50 robot arms at its Factory Zero EV plant in Detroit, sparking union backlash as over 1,300 workers remain laid off. The move highlights tensions between automation and labor, with UAW leaders accusing GM of prioritizing profits over workers.

NVIDIA announces Halos for Robotics, the industry's first full-stack safety system for physical AI, built on over 18,600 engineering years of autonomous vehicle safety development. Agility Robotics is the first to adopt the system for its humanoid robots in industrial environments.

NVIDIA's ENPIRE framework, developed with CMU and UC Berkeley, uses AI coding agents to autonomously train robots for high-precision physical tasks like GPU installation, achieving a 99% success rate through a closed feedback loop and real hardware trials.

University of Michigan Robotics shares free open-source course materials including lectures, textbooks, and projects from their top robotics program, covering topics from computational linear algebra to autonomous systems.

ShotcreteDepth is a bi-modal dataset of stereo RGB and LiDAR data from construction environments, designed to support research in depth perception under challenging conditions. The dataset includes 11,252 samples with 220 annotated, and is accompanied by a lightweight annotation tool.

ENPIRE is a framework that enables coding agents to autonomously improve robot manipulation policies through a real-world feedback loop, achieving 99% success on dexterous tasks like pin insertion and zip tie cutting.

Robotics teams are rebuilding the data stack from scratch to overcome the 'data layer tax' that slows down iteration and scaling in robot learning, as existing infrastructure doesn't handle multi-rate and multimodal data.

PolicyTrim is a reinforcement learning-based post-training framework that improves action chunk utilization by 3× and reduces physical execution steps by 51.4% in Vision-Language-Action models, delivering up to 5.83× deployment speedup.

Elon Musk tweets about a future where millions of humanoid robots can build Manhattan in months, envisioning abundance by 2045 and beyond.

An analysis of the software stack behind autonomous robots, breaking down the components from perception to cloud support, and highlighting that most tools are open-source.

This paper introduces Reward as an Agent and DynDiff-GRPO to address reward hacking and limited exploration in reinforcement learning for embodied world models, achieving significant accuracy gains.

A clarification post noting that a robot face shown is actually a real robot, unlike a previous misleading post.