Techniques are disclosed for a combined machine learning (ML) model that may generate tracking confidence metrics associated with tracking and/or classification of an object. The technique may include obtaining a tracking. Tracking, which may include object detection from one or more sensor data types and/or pipelines, may be input into a machine learning (ML) model. The model may output a tracking confidence metric and classification. In some examples, if the tracking confidence metric does not meet a threshold, the ML model may result in suppression of the tracking output of the planning component of the autonomous vehicle.
本发明公开了一种用于组合的机器学习(ML)模型的技术,该模型可以生成与对对象的跟踪和/或分类相关联的跟踪置信度度量。该技术可以包括获得跟踪。可以包括来自一种或多种传感器数据类型和/或管线的对象检测的跟踪可以被输入到机器学习(ML)模型中。该模型可以输出跟踪置信度度量和分类。在一些示例中,如果跟踪置信度度量不满足阈值,则ML模型可能会导致对自主车辆的规划组件的跟踪输出的抑制。
Combined tracking confidence and classification model
组合的跟踪置信度和分类模型
2022-09-30
Patent
Electronic Resource
Chinese
IPC: | B60W CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION , Gemeinsame Steuerung oder Regelung von Fahrzeug-Unteraggregaten verschiedenen Typs oder verschiedener Funktion / G06N COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS , Rechnersysteme, basierend auf spezifischen Rechenmodellen / G06V / G08G Anlagen zur Steuerung, Regelung oder Überwachung des Verkehrs , TRAFFIC CONTROL SYSTEMS |
Laser-Based Vehicles Tracking and Classification Using Occlusion Reasoning and Confidence Estimation
British Library Conference Proceedings | 2008
|