China's Baidu launches AI model simulating road environments for robotaxis
BIDU•Baidu launched Apollo World, an AI system for simulating environments to train robotaxis and predict how they evolve. It can generate up to 100,000 km of simulated driving daily, and Baidu says it is being used to train Apollo Go's robotaxi fleet.
1. Simulated driving at scale
Baidu launched Apollo World, an AI system designed to simulate real-world environments for robotaxi training and predict how they evolve. The company says the system can generate up to 100,000 km of simulated driving daily, equivalent to about three years of real-world testing by a safety-driver-equipped autonomous vehicle.
2. Supporting Apollo Go
Apollo World can produce real-time multimodal data, including camera and lidar observations, for self-driving systems to sense and navigate simulated environments. Baidu said it is already being used to train Apollo Go's robotaxi fleet, which is to be operated globally.
3. Costs remain a challenge
Chinese robotaxi firms have largely pursued asset-light expansion by supplying technology and partnering with automakers, ride-hailing platforms and local fleet operators. Autonomous mobility expert Frank McCleary said model training and vehicles still involve capital expenditure. Morgan Stanley has estimated China's robotaxi fleet could grow to 1 million vehicles by 2035.
