Robots and autonomous systems: manipulators, humanoids, drones and robot learning.
The open-source Rust framework supports orthogonal, polygonal, and curvilinear geometries with configurable parameters and probability distributions for room placement and obstacles.
The company plans to significantly increase its robot fleet to automate the most physically demanding tasks, aiming to protect over 1.5 million human workers from fatigue and injury.
Researchers remain skeptical that current language models can quickly teach robots to control the physical world.
A meta-agent analyzes task trajectories to diagnose failures and revise tools, achieving a 62% success rate across 42 RoboDojo tasks without updating model weights.
Founded in 2024, Mecka AI launched a program to collect human motion data for training humanoid robots, following a similar model used by data-labeling companies for LLMs.
IEEE Spectrum's white paper introduces the HiPHI dataset, designed to address the data gap hindering humanoid robot learning in tasks like carrying and pushing.