Artificial Intelligence in the Field of Driverless Cars

Nowadays, new technologies led by artificial intelligence and big data are booming, and have been widely used in many fields such as medical, transportation, military, and industrial. Automobile driving is one of the main application directions of artificial intelligence in the field of transportation. The purpose of this article is to explore the application of artificial intelligence technology in the field of driverless cars, with a view to bringing precision and real-time improvement to pedestrian detection in driverless systems. In this paper, based on the verification model and the pedestrian recognition algorithm based on the recognition model, a new recognition algorithm combining two types of models is proposed. After optimization and testing on the Market1501 pedestrian re-identification data set, it is concluded that the multi-camera test mAP of this algorithm is 72.55%, which greatly improves the detection accuracy compared with other mainstream re-identification algorithms. The research in this paper basically solves the problems of low accuracy, poor real-time performance and discontinuous cross-camera detection, which is beneficial to the successful application of pedestrian detection in unmanned driving.