News
Google DeepMind Pushes Robots Toward Real-World Autonomy With Embodied Reasoning Upgrade
Google DeepMind has released Gemini Robotics-ER 1.6, an upgrade to its embodied reasoning model for robotics.
👁️ Perception Systems
Tools and platforms that give machines the ability to sense, interpret, and act on the world, from sensor fusion to robot foundation models.
News
Pony.ai Introduces Self-Improving World Model for Autonomous Driving
Pony.ai announced PonyWorld 2.0, a new version of its world model that serves as the core training system behind its autonomous driving stack. The company describes the update as a step toward a “self-improving approach” to development, where the system contributes to how it is trained.
News
AGIBOT Pushes World Models Into Interactive Simulation With Genie Envisioner 2.0
AGIBOT has debuted a major update to Genie Envisioner, its world models platform. The update moves the system from describing the world to running it as a simulation. It builds on the World Action Model, linking state, action, and outcome so changes unfold step by step.
News
RealSense Pushes Humanoid Robots Closer to Safe, Autonomous Movement in Human Environments
At NVIDIA GTC, RealSense showed a humanoid robot moving autonomously in a real-world environment, built in collaboration with LimX Dynamics. The demonstration focused on how perception supports safe operation for robots working around people.
News
Niantic Spatial Teams With Coco Robotics on Navigation Infrastructure for Delivery Robots
Niantic Spatial announced a partnership with Coco Robotics aimed at improving navigation systems used by autonomous delivery robots. The agreement positions Niantic Spatial as a core infrastructure partner, supplying spatial AI capabilities and a visual positioning system
News
Harbinger Acquires Phantom AI and Secures ADAS Licensing Deal With ZF
Harbinger announced it has acquired Phantom AI and entered into a licensing agreement with ZF’s passenger car Advanced Driver Assistance Systems division. ZF will license Phantom AI’s computer vision software for its passenger car ADAS products
News
Looper Robotics Pushes Industry Beyond Stunt Robotics With Debut of Spatial AI Camera
Looper Robotics has debuted Insight9, described as the world’s first Autonomous Spatial AI Camera, which aims to "democratize high-end robot perception." The product will be made available globally on Kickstarter as part of the company’s
News
ZaiNar Raises $100 Million to Turn Existing Wireless Networks Into a Sensing Platform for Physical AI
ZaiNar announced it has raised more than $100 million at a valuation exceeding $1 billion as it publicly launches after nine years in stealth with a platform for Physical AI. The company says its technology converts existing wireless networks into a sensing system
News
Waymo Introduces 6th-Generation Driver as Foundation for Scaled Expansion
Waymo is beginning fully autonomous operations with its 6th-generation Driver. The company says the system serves as the primary engine for its next era of expansion. Waymo's latest Driver builds on nearly 200 million fully autonomous miles driven across more than 10 major cities and freeways.
News
Ouster Acquires StereoLabs to Add Vision to Its Lidar Business
Ouster said it has completed its acquisition of StereoLabs. Ouster sells digital lidar products. StereoLabs develops stereo cameras and 3D perception software and has been operating since 2010.
Editorial
How Computer Vision Is Teaching Machines to Read the Wild
Camera traps capture rare wildlife moments, but millions of images need sorting. Computer vision tools like MegaDetector and SpeciesNet turn chaos into data, helping scientists track species, find rare animals, and speed conservation work worldwide.
News
U.S. Ski & Snowboard and Google Cloud Test Smartphone-Based AI Motion Analysis for Training
U.S. Ski & Snowboard and Google announced a joint effort to build an AI video-analysis tool on Google Cloud aimed at elite ski and snowboard athletes.