The main goal of this report is to build a detector of all objects present in the surrounding of an autonomous vehicle on roads. We achieved this goal by means of a deep learning algorithm called YOLO (You Only Look Once) in the field of computer vision. The basic concepts in object detection as well as the core of the YOLO algorithm are recalled in this study. Our machine learning modeling operations consist of the three following steps. Firstly, we collect a large dataset of images containing 35 classes of traffic scene objects. Secondly, we annotate the collected images in the YOLO format. Thirdly, we feed a model from the YOLO family with the annotated dataset in order to estimate its parameters. The obtained model has pretty good predictive performance and can be used to extract in real-time all information associated with the external driving environment from videos taken by cameras embedded in an autonomous vehicle. Traffic scene elements extracted by such a detector can act as covariates in reliability analysis of automated driving systems consisting to check whether a safety requirement is satisfied in an operational design domain.