Gemini Robotics 2 demonstrates full-body humanoid control
Apollo 2 demonstrations combined walking, crouching, reaching, and dexterous manipulation under one embodied model.
Apollo 2 demonstrations combined walking, crouching, reaching, and dexterous manipulation under one embodied model.
A systems framework connects AI safety, control, cybersecurity, dependable computing, human factors, and lifecycle assurance.
An exploratory mobile-robot study reports spatial memory and action consequence reasoning from a general multimodal model.
A diffusion-transformer approach jointly models robot embodiment, states, and actions to optimize hardware for target motions.
RoboBRIDGE proposes modular interfaces that connect learned policies with robust real-world robot execution.
The World Action Planner uses action-conditioned world models to improve generalizable robot decision-making.
A cross-embodiment method seeks reusable behavior representations spanning different robot bodies and control spaces.
LabEvolver explores safe, grounded adaptation for embodied agents operating in wet-lab environments without conventional retraining.
A criticality-aware framework seeks to guide self-evolving embodied learning toward safer and more consequential experiences.
RedFlow redirects detected failures into action-level corrections for flow-matching vision-language-action policies.