Embodied AI
AI systems that learn, reason and act through interaction with physical environments, sensors and actuators.
Knowledge base / Concepts
Robot embodiment, explained.
A growing reference for the ideas connecting artificial intelligence with physical machines.
Physical Intelligence
The broader pursuit of intelligent behavior in the physical world, combining learning, perception, planning and control.
Robot Foundation Models
General-purpose models intended to transfer knowledge across tasks, environments or robot embodiments.
Vision-Language-Action
Systems that connect visual observations and language instructions to physical actions.
World Models
Learned models that represent aspects of an environment and support prediction, planning or decision-making.
Sim-to-Real
Techniques for transferring behaviors learned in simulation to real robotic hardware.
Robot Manipulation
The perception, planning and control of physical interaction with objects and environments.
Dexterous Manipulation
Fine-grained manipulation using capable end-effectors, hands and coordinated control.