Baidu unveils Apollo World, a physical AI world model for autonomous driving simulation
Baidu_Inc · x · 2026-11-04
At Baidu AI Day in Shenzhen, Baidu unveiled Apollo World, a world model for physical AI that turns complex road interactions and rare driving scenarios into repeatable, controllable simulations for training and testing autonomous driving systems.
Key capabilities:
- AI traffic agents that make their own decisions and react as autonomous vehicles slow down, change lanes, or yield
- Real-time simulated sensor data (camera, LiDAR, etc.), letting vehicles sense and respond to a virtual world that changes with their actions
- Complete simulated rides between any two locations in a virtual city, from accepting a request to drop-off
- A technology stack extensible to robotics and other physical AI applications
Apollo World is already part of Apollo Go's testing and validation, with insights from virtual trips feeding back into real-road driving. Slides shown at the event cite a safety decision cycle within 1/30 of a second.
Related event: Baidu Unveils Apollo World Model for Controllable Driving Simulation(2 posts)→
More from Embodied
- Autonomous to roll out Harness v2 agent command center update next week — dee_hw · 2026-11-04
- Two Humanoid Robots Log 560K Records on Korean Injection Molding Line — CyberRobooo · 2026-11-04
- Gaussian splatting game built with Playcanvas: real2sim as a training ground for robotics — willeastcott · 2026-11-04
- Multi-fingered robotic hand demos coordinated bends, real dexterity still unproven — burhop · 2026-11-04
- This robot runs an entire 3D printing farm unattended, print after print — lukas_m_ziegler · 2026-11-04
- Tsinghua & NVIDIA propose PrismWAM, splitting semantics, physics and actions in world models — jiqizhixin · 2026-11-04