39 datasets found

Learning language-conditioned robot behavior from offline data and crowd-sourced annotation.

Open X-Embodiment: Robotic learning datasets and RT-X models

Aligning robot behavior with human preferences is crucial for deploying embodied AI agents in human-centered environments. A promising solution is interactive imitation learning...

The dataset used in the paper is RLBench, a standard benchmark for vision-based robotics which has been shown to serve as a proxy for real-robot experiments.

The Realworld Kitchen dataset contains four sub-tasks, namely, opening the oven, grasping a cloth, closing a drawer, and turning on a light.

The Franka Kitchen environment is a challenging long-range control problem, which involves a simulated 9-DOF Franka Emika Robot in a kitchen setting.

Human demonstration videos are a widely available data source for robot learning and an intuitive user interface for expressing desired behavior.

Scaling robot learning with semantically imagined experience.

The YCB object and model set is a benchmark for manipulation research, consisting of 15 object categories and 3D models.

The dataset is used for training autonomous quadrotor navigation policies for direct, real-world deployment.

We propose a novel methodology to learn neural contractive dynamical systems, where our neural architecture ensures contraction, and hence, global stability.

We propose a novel methodology to learn neural contractive dynamical systems, where our neural architecture ensures contraction, and hence, global stability.

We propose a novel methodology to learn neural contractive dynamical systems, where our neural architecture ensures contraction, and hence, global stability.

We propose a novel methodology to learn neural contractive dynamical systems, where our neural architecture ensures contraction, and hence, global stability.

Stability guarantees are crucial when ensuring a fully autonomous robot does not take undesirable or potentially harmful actions. We propose a novel methodology to learn neural...

The Sparse Diffusion Policy (SDP) framework, which integrates Mixture of Experts (MoE) layers into the diffusion policy.

PartNet Mobility and GAPart-Net are datasets of articulated objects with their part configurations.

Gen2Sim generates 3D assets from object-centric images using image diffusion models and predicts physical parameters for them using LLMs.

D4RL datasets for maze2d-umaze, maze2d-medium, maze2d-large, antmaze-umaze, antmaze-medium, antmaze-large, parking, soccer-sim, and soccer-physical tasks