Robotics foundation model
A general-purpose pre-trained model capable of driving different robotic morphologies without retraining.
A machine learning model pre-trained on large corpora of multimodal data (video, sensors, text) and capable of driving robots with varied morphologies (humanoids, quadrupeds, articulated arms) without being retrained for each body form. Cross-morphology generalisation is the central promise of these models.
Articles about “Robotics foundation model”
- There is no longer a shortage of robots, there is a shortage of Unitree shares
- LG announces a bipedal humanoid for Q1 2027, running on Jetson Thor and Isaac GR00T
- Why does the US block kick in at 200 kbit/s?
- The surgical robot is controlled by a surgeon
- Five times "autonomous" in six days: the word that doesn’t weigh the same twice
Related terms
Join the discussion Soon
A question, a disagreement, hands-on feedback on this term? The Botoide community forum is coming soon.