百度发布Apollo World世界模型，用仿真环境训练自动驾驶车辆
来源标题：Baidu launches Apollo World model to train robotaxis in simulated environments
百度11月3日发布Apollo World世界模型，可模拟交互式道路环境用于训练和测试自动驾驶车辆，每天能生成最多10万公里仿真驾驶数据。该系统已用于支持Apollo Go的全球Robotaxi开发，能实时生成摄像头和激光雷达数据，虚拟车辆与行人会对自动驾驶车辆的动作做出反应。
以下正文同步自 动点科技 TechNode，版权归原站所有，已转换为易读排版。
Baidu unveiled Apollo World on November 3, a world model that simulates interactive road environments to train and test autonomous vehicles. The company said the system is already being used to support Apollo Go’s global robotaxi development and can generate up to 100,000 kilometers of simulated driving per day. Apollo World produces real-time camera and lidar data, while virtual vehicles and pedestrians respond to the autonomous vehicle’s actions. This allows driving systems to practice complete trips through changing traffic conditions and repeatedly test difficult scenarios. The launch comes as robotaxi developers seek to reduce the cost and time required to adapt their technology to new markets. [Reuters]
正文由 探How 从来源站点 RSS 同步，内容未经改写；遇到排版缺失或需要图片、视频时请以原文为准。