Tag
143 articles
London robotics startup Humanoid has raised $152 million in a Series A round, pushing its valuation above $1 billion. The funding includes participation from Prime Movers Lab and Bosch, which will manufacture its wheeled robots at scale.
NVIDIA has released Cosmos 3 Edge, a 4-billion-parameter open world model designed to run entirely on-device, enabling robots to reason and act autonomously in real-time environments.
How Xiaomi's robot learned to move by watching humans, and why more data beats bigger models in AI.
Foundation Future Industries, backed by Eric Trump, is reportedly developing military robotics capabilities, raising questions about the future of autonomous weapons systems.
Learn how brain-computer interfaces (BCIs) allow people to control robots with their thoughts using a lightweight headset and AI technology.
German startup microagi has raised $55 million to train robots using human activity footage, marking the largest seed round for a German company. The company aims to revolutionize robot learning by leveraging real-world human behavior data.
Japan is building a 140MW AI factory for robots in collaboration with Nvidia, marking a major step toward integrating AI into physical systems. The facility will house 13,750 Nvidia Vera CPUs and 27,500 Rubin GPUs.
Nvidia has enlisted most of Japan’s industrial robotics sector into its Cosmos Coalition, a move aimed at advancing AI-driven robotics through open-world models.
This article explains the technical and strategic reasoning behind Walden Robotics' decision to build legless humanoid robots for factory use, focusing on control systems, AI integration, and industrial robotics design.
Learn how Boston Dynamics' Spot robot uses artificial intelligence to deliver packages from trucks to doorsteps, and why this technology matters for the future of delivery services.
Mistral AI introduces Robostral Navigate, an 8B model that enables robots to navigate complex environments using only a single RGB camera, without requiring LiDAR or depth sensors.
Learn to build a simplified world model inspired by China's Orca system that predicts abstract world states without requiring labeled action data, demonstrating how unsupervised learning can reduce data requirements in robotics.
