110 claims over the last 90 days
Robust.AI is progressing well since its 2019 founding
Deploying robots to real-world applications requires a long development timeline
Gemini Robotics ER 2 enables robots to reason, collaborate, and solve real-world tasks
Gemini Robotics ER 2 represents a step change in video understanding, tool orchestration, and multi-robot collaboration for robotic applications
Gemini Robotics 2 enables robots to reason through every movement to manage tasks that weren't possible before, like tying delicate knots
Gemini Robotics 2 allows robots to team up to solve complex workflows
World Labs is building virtual worlds that can train robots as part of their spatial intelligence vision
Universal physical laws govern spatiotemporal dynamics regardless of the actor
State-of-the-art action-conditioned video models are restricted to single robot embodiments and cannot leverage heterogeneous video data
CLAP is capable of being trained on diverse, internet-scale videos across human and robotic agents for cross-embodiment action-conditioned video generation
CLAP approaches or surpasses state-of-the-art single-embodiment video models in challenging environments like DROID
CLAP establishes a novel paradigm for training single-embodiment video world models through few-shot adaptation
CLAP delivers the most comprehensive suite of action-conditioned video world models to date spanning diverse action-conditioning spaces and robot morphologies
Multi-modal Large Language Models used for task planning in industrial human-robot collaboration inherently lack an understanding of system states and do not track state transitions, leading to hallucinated actions that deviate from the intended goal
Generating action plans in natural language limits the plans to a high level and introduces ambiguity in action execution
State-aware Task Estimator and Planner (STEP) outperforms state-of-the-art by 32.8% in action executability and 14.8% in final-state error for robot assembly tasks
Prompting Multi-modal LLMs to explicitly estimate system state and predict state transitions ensures task-convergent planning in human-robot collaboration
Google launched Gemini Robotics 2, one AI system designed to control everything from robot arms to full humanoids
Embodied intelligence faces a fundamental data bottleneck because existing datasets fragment the full perception-action loop across viewpoints, modalities, or spatial scales
The Ambient Capture Engine (ACE) can transform real home environments into spatially calibrated, temporally synchronized recording studios that capture the full perception-action loop as a unified multisensory stream