Tag
This paper formalizes the sim-to-real gap for foundation model agents as a Markov Decision Process problem, proposing a unified research agenda to adapt classical solutions like domain randomization for improving agent robustness and reliability in real-world deployment.
OASIS is a simulation-data-driven framework for humanoid loco-manipulation that uses 3D generative models and hierarchical visuomotor policies. It achieves better zero-shot performance than real-robot training by leveraging domain randomization in simulation.
This paper introduces Domain-Randomized Instance Set (DRIS), a method that simultaneously represents multiple randomized instances to improve sim-to-real transfer for dexterous manipulation. It demonstrates zero-shot transfer on a challenging reactive catching task with a flat plate end-effector, requiring no real-world fine-tuning.
OpenAI developed a robot hand capable of solving a Rubik's Cube using a novel technique called Automatic Domain Randomization (ADR), which progressively increases simulation difficulty to enable effective transfer of learned behaviors from simulation to the real world.
OpenAI describes challenges with conventional RL on robotics tasks and introduces Hindsight Experience Replay (HER), a new RL algorithm that enables agents to learn from binary rewards by reframing failures as intended outcomes, combined with domain randomization for sim-to-real transfer.
Researchers explore a data generation pipeline using domain randomization and procedurally generated objects to train a deep neural network for robotic grasp planning. The proposed autoregressive model achieves >90% success on unseen objects in simulation and 80% in the real world, despite being trained only on random simulated objects.
OpenAI open-sources mujoco-py, a high-performance Python library for robotic simulation using the MuJoCo engine, featuring ~40x speedup with headless GPU rendering and VR interaction support.
OpenAI demonstrates that domain randomization—randomly varying colors, textures, lighting, and camera settings in simulated training data—enables deep learning models to effectively transfer from simulation to real-world robotic spam detection tasks without retraining from scratch.