Vehicle navigation path optimization based on complex networks

Abstract

Vehicle navigation path optimization, an essential means to prevent and alleviate traffic congestion, assists users in finding optimal routes from origin to destination based on acquired traffic information. This paper proposes a vehicle navigation path optimization approach that incorporates complex networks. Initially, a complex network-based multi-objective optimization model is developed to address total travel time and cost objectives. Subsequently, the Non-dominated Sorting Genetic Algorithm II (NSGA-II) is enhanced by integrating a machine learning approach and designing a competitive selection operator, along with crossover and mutation operators based on hierarchical clustering, to create a multi-objective vehicle navigation path optimization algorithm. Finally, case studies validate the model and algorithm’s effectiveness. Experimental results demonstrate the superiority of the proposed machine learning and NSGA-II hybrid algorithm over traditional NSGA-II and NSGA-III. This research achieves rational and balanced distribution of traffic flow across road segments by appropriately guiding vehicles, thereby improving traffic network efficiency.