Generalizing from simulation

OpenAI Blog · 8 years ago

Researchers developed methods to train robot controllers in simulation that can then operate effectively on physical robots while adapting to unexpected environmental changes. The systems now use closed-loop control rather than open-loop control, allowing real-time feedback and adjustment during task execution. This enables robots trained entirely in simulation to handle dynamic conditions without retraining when deployed on actual hardware.