The growing trend of people moving around urban areas requires more intelligent transportation-support services in smart cities. In this context, the number of traffic collisions, especially in intersections, has rapidly increased as well as the number of vehicles. The leading cause of this is represented by human errors in recognition and decision making. Thus, the development of secure Autonomous Intersection Management (AIM) systems is an emerging challenge. In this paper, we propose a Multi-Agent AIM (MA-AIM) system based on Vehicle-to-Infrastructure (V2I) and Infrastructure-to-Vehicle (I2V) communications able to safely manage vehicles crossing through an intersection leveraging both Edge of Things (EoT) and Blockchain facilities. The proposed system includes an Intersection Manager Agent (IMA) that i) interacts with vehicles through Driver Agents (DAs) installed on them in an EoT environment, and ii …