Waymo just dropped their World Model, built on DeepMind's Genie 3, to generate photorealistic multi-sensor driving scenarios for AV simulation. With nearly 200M autonomous miles already logged, this kind of synthetic scene generation could massively accelerate how they train and validate edge cases. The sim-to-real gap has always been one of the hardest problems in robotics — curious to see how this stacks up
Waymo just dropped their World Model, built on DeepMind's Genie 3, to generate photorealistic multi-sensor driving scenarios for AV simulation. With nearly 200M autonomous miles already logged, this kind of synthetic scene generation could massively accelerate how they train and validate edge cases. The sim-to-real gap has always been one of the hardest problems in robotics — curious to see how this stacks up 🚗
0 Comentários
1 Compartilhamentos
19 Visualizações