OpenDriveLab
ReSim: Reliable World Simulation for Autonomous Driving

ReSim: Reliable World Simulation for Autonomous Driving

ReSim is a driving world model that enables Reliable Simulation of diverse open-world driving scenarios under various actions, including hazardous non-expert ones. A Video2Reward model estimates the reward from ReSim's simulated future.

FreeTacMan: Robot-free Visuo-Tactile Data Collection System for Contact-rich Manipulation

FreeTacMan: Robot-free Visuo-Tactile Data Collection System

A human-centric and robot-free visuo-tactile data collection system for high-quality and efficient robot manipulation.

UniVLA: Learning to Act Anywhere with Task-centric Latent Actions
RSS 2025

UniVLA: Learning to Act Anywhere with Task-centric Latent Actions

A unified vision-language-action framework that enables policy learning across different environments.

AgiBot World Colosseo: A Large-scale Manipulation Platform for Scalable and Intelligent Embodied Systems
IROS 2025

AgiBot World

World's First Large-scale High-quality Robotic Manipulation Benchmark.

DriveLM: Driving with Graph Visual Question Answering
ECCV 2024 Oral

DriveLM: Driving with Graph Visual Question Answering

Unlocking the future where autonomous driving meets the unlimited potential of language.

Planning-oriented Autonomous Driving
CVPR 2023 Best Paper Award

Planning-oriented Autonomous Driving

UniAD: The first comprehensive framework that incorporates full-stack driving tasks.

We are searching for talents from all over the world. Are you looking for opportunities? Don't hesitate to contact us via [email protected] or Dr. Hongyang Li.

  • 具身智能研究員 / 自動駕駛研究員 / 機器人硬件工程師 / 科研助理 / 生態合作助理 【更多詳情】
  • Ph.D. student / Research Assistant / Postdoc / etc. in Hong Kong and Shanghai
  • Full-time employee and Intern (international are welcome)
Representative work published at top-tiered venues.
A unified vision-language-action framework that enables policy learning across different environments.
RSS 2025
UniAD: The first comprehensive framework that incorporates full-stack driving tasks.
A novel generalist policy that leverages latent action representations to maximize data utilization, demonstrating predictable performance scaling with increased data volume.
IROS 2025
blog
huggingface
bilibili
In this survey, we provide a comprehensive analysis of more than 270 papers on the motivation, roadmap, methodology, challenges, and future trends in end-to-end autonomous driving.
IEEE-TPAMI 2024Survey
Unlocking the future where autonomous driving meets the unlimited potential of language.
ECCV 2024 Oral
dataset
page
huggingface
AgiBot World
AgiBot World
Manipulation
  • Cutting-edge Sensor and Hardware Design.
  • Wide-spectrum of Scenario Coverage.
  • Quality Assurance with Human-in-the-loop.
OpenDV
OpenDV
GenerationWorld Model
  • The largest driving video dataset to date, containing more than 1700 hours of real-world driving videos.

Challenge 2025

Autonomous Grand Challenge

Honolulu

Honolulu

Workshop at ICCV 2025

Learning to See: Advancing Spatial Understanding for Embodied Intelligence

October 19, 2025