News News
Last Updated: 2024-10-19 17:25:23
2024-10
We have released OmniGibson v1.1.0, with streamlined installation, parallized environments, primitive skills, and data collection pipeline2024-09
Our recent paper, Automated Creation of Digital Cousins for Robust Policy Learning, was accepted at CoRL20242024-06
Our recent paper, BEHAVIOR Vision Suite: Customizable Dataset Generation via Simulation, was accepted at CVPR2024 [Highlight]
2024-03
We have released OmniGibson v1.0.0, with all 1000 pre-sampled tasks, 50 large-scale scenes, and 10000 unique object assets
2024-03
We presented BEHAVIOR-1K at NVIDIA’s GTC conference
2023-08
We have released OmniGibson v0.2.1, with improved stability and refined assets
2023-04
We have released OmniGibson v0.2.0, with more assets: 600 pre-sampled tasks, 7 new scenes, and many new objects
2022-12
We have released OmniGibson v0.1.0, with significantly improved stability, performance, and ease of installation
2022-12
We have released OmniGibson v0.0.1, a hyper-realistic simulation platform for accelerating embodied AI research
2022-11We have released robosuite v1.4, which supports the newest DeepMind mujoco bindings
2022-10We hosted a workshop on Benchmarking Embodied AI Solutions in Natural Tasks at ECCV2022
2022-09Our recent paper, BEHAVIOR-1K: A Benchmark for Embodied AI with 1,000 Everyday Activities and Realistic Simulation, was accepted at CoRL2022 [Best Paper Nominee]
2022-06We hosted a workshop on Overlooked Aspects of Imitation Learning: Systems, Data, Tasks, and Beyond at RSS2022
2022-01Our recent paper, OSCAR: Operational Space Control for Adaptive Robustness, was accepted at ICRA2022
2021-10We have released robosuite v1.3, with support for additional renderers and vision modalities
2021-09We have released robomimic v0.1, a framework for robot learning from demonstration
2021-09Our recent paper, What Matters in Learning from Offline Human Demonstrations for Robot Manipulation, was accepted at CoRL2021 [Oral]
2021-09Our recent paper, Error-Aware Imitation Learning from Teleoperation Data for Mobile Manipulation, was accepted at CoRL2021
2021-07Our recent paper, iGibson 1.0: a Simulation Environment for Interactive Tasks in Large Realistic Scenes, was accepted at IROS2021
2021-07We hosted a workshop on End-to-end GPU-accelerated Learning and Control for Robotics with Isaac Gym at RSS2021
2021-03Our recent paper, Learning Multi-Arm Manipulation Through Collaborative Teleoperation, was accepted at ICRA2021 [Best Multi-Robot Systems Paper Finalist]
2021-03We have released robosuite v1.2, with support for sensor and dynamics randomization
2020-12We have released robosuite v1.1, with refactored infrastructure to streamline environment prototyping
2020-09We have released robosuite v1.0, a modular simulation framework and benchmark for robot learning