MAAVSL-HIT2FS: Multi-Agent Based Architecture for Variable Speed Limit Decision making based on Hierarchical Interval Type 2 Fuzzy System

Abstract

Actual urban road networks are characterized by a high level of dynamism, uncertainty, and more frequent situations of traffic congestion/jams which can lead to more complex traffic control and management. Traffic congestion is influenced by several aspects such as peak hours, accidents, weather, and road works. To deal with such a situation, in this research work, we propose a multi-agent distributed architecture to the Cooperative Intelligent Transport System (C-ITS) based on hierarchical Interval Type-2 fuzzy logic (HIT2FL) for Variable Speed Limit (VSL) decision-making. Our main goal is to enhance road safety and traffic efficiency by using realtime information and taking into consideration contextual factors as well as warning events to propose the appropriate speed limit value. We validate our approach using SUMO as a microscopic traffic-based simulator and Jade platform for multiagent systems. Our approach outperforms average delay, mean speed, mean travel time and vehicles flow and its effectiveness is verified by comparing the outcomes with existing methods including HIT2FLS as well as a comparison with the Original Traffic Trace (OTT).