RAPID: Integrating AI and Multi-Agent Systems for Enhanced Traffic Management Framework with YOLOv9

Traffic accidents pose a major risk to human life and have a considerable economic impact worldwide. In this study, we present the RAPID framework, an innovative approach that integrates Artificial Intelligence (AI) and Multi-Agent Systems (MAS) to enhance traffic management. The core of the RAPID framework is the YOLOv9 model, a “state-of-the-art” deep learning (DL) algorithm for real-time object detection and classification, which we employ to detect and classify traffic accidents accurately. The RAPID framework addresses key challenges in modern traffic management, including data heterogeneity, scalability and real-time processing. Our system leverages a multi-layered architecture: data is acquired from various sources, including traffic cameras and CCTV and then transmitted to a cloud data center for storage and processing. Within the cloud infrastructure, the data undergoes pre-processing and is analyzed by the YOLOv9 model. The MAS coordinates various components of the framework, distributing tasks among specialized agents responsible for data acquisition, pre-processing, model training and real-time detection. We evaluated the performance of the RAPID framework using a dataset of 3000 labeled traffic accident images. The YOLOv9 model demonstrated 95.4% accuracy, 93.8% precision, 92.6% recall and 93.2% F1-score. The experimental results indicate that the RAPID framework significantly improves traffic accident detection and response times compared to traditional systems, highlighting its potential for real-world applications. The RAPID framework contributes to the field of intelligent transportation systems (ITS) by providing a robust, scalable and efficient solution for real-time traffic management.