In Pictures: Baidu unveils Apollo World AI model to train autonomous driving systems in virtual traffic

Apollo World simulates complex traffic scenarios to train autonomous driving systems without real-world risk.

Photo by Jan Wright on Pexels

Baidu says a single day of virtual testing in Apollo World equals roughly three years of real-world driving.

Photo by HRK Gallery on Pexels

The world model generates interactive AI traffic agents to stress-test self-driving strategies before deployment.

Photo by Acres of Film on Pexels

Apollo World aims to cut costs and deployment risks for autonomous driving development.

Photo by Mike Bird on Pexels

Virtual validation helps Baidu's Apollo Go robotaxi service scale testing across global markets.

Photo by Sami Aksu on Pexels

Real-world traffic complexity, such as in Guangzhou, informs the simulated environments Apollo World recreates.

Photo by zhen tang on Pexels

Interactive AI agents in Apollo World mimic diverse vehicles and pedestrian behavior found in dense cities.

Photo by Blackcurrant Great on Pexels

Baidu positions Apollo World as a core tool for validating autonomous driving decisions at scale.

Photo by Антон Залевський on Pexels

The model is already supporting global testing of Baidu's Apollo Go robotaxi service.

Photo by zhen tang on Pexels