首页> 外国专利> Sound source detection and location identification for autonomous vehicles

Sound source detection and location identification for autonomous vehicles

机译:自动驾驶汽车声源检测与位置识别

摘要

PROBLEM TO BE SOLVED: To recognize real-time sound source and position identification of an obstacle or an object around an own vehicle. SOLUTION: Audio data is received from a plurality of audio sensors attached to an autonomous driving vehicle (ADV). Audio data includes sounds captured by audio sensors and emitted from one or more sound sources. In this method, a plurality of sound source information is further determined based on the received audio data. Each source information includes a confidence score associated with the presence of a particular sound. The method further generates a data representation for reporting the presence or absence of a particular sound in the operating environment of the ADV. The data representation includes determined sound source information. The machine learning algorithm is subsequently trained using the received audio data and the generated data representation to recognize a specific sound source in real time during autonomous driving of the ADV. [Selection diagram] FIG. 7
机译:需要解决的问题:实时识别声源,并对车辆周围的障碍物或物体进行位置识别。解决方案:从连接到自动驾驶车辆(ADV)的多个音频传感器接收音频数据。音频数据包括由音频传感器捕获并从一个或多个声源发出的声音。在该方法中,基于接收到的音频数据进一步确定多个声源信息。每个源信息包括与特定声音存在相关的置信度分数。该方法还生成用于报告ADV操作环境中特定声音的存在或不存在的数据表示。该数据表示包括确定的声源信息。随后使用接收到的音频数据和生成的数据表示来训练机器学习算法,以在图7的ADV.[选择图]的自动驾驶期间实时识别特定声源

著录项

相似文献

  • 专利
  • 外文文献
  • 中文文献
获取专利

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号