Spatial–Temporal Resource Assignment for Variable Guidance Lanes under the Intelligent Vehicle Infrastructure Cooperative Environment

Due to the increased frequency and magnitude of changes in urban road traffic demand, temporal resource optimization alone is no longer effective in alleviating congestion at intersections. A spatial resource optimization mode based on variable guiding lanes has caused some confusion to drivers, making it difficult to fully improve spatial resource optimization efficiency. This paper proposes a multiobjective optimal spatial and temporal resource assignment model based on the intelligent vehicle infrastructure cooperative system (I-VICS) and variable guidance lane (VGL). The model constrains the number of lanes, signal timing, traffic demand, evaluation indicators, communication conditions, and type of variables, aiming to maximize green signal utilization and minimize lane occupancy ratio. Through the introduction of a speed guidance strategy and a signal-coordinated optimization model, demand-responsive control coordinated with signals is achieved. The assignment model is solved by combining a fuzzy inference system and the nondominated sorting genetic algorithm with the elite strategy (NSGA-II). Finally, the model is integrated with the solving method into an intelligent control process for practical application. The interaction of information between the onboard units and roadside units eliminates information barriers for drivers. Simulation results demonstrate that the proposed solution can significantly reduce the average vehicle delay by 25.34% and average traveling time by 38.59%. Furthermore, under oversaturated conditions during peak hours, both green signal utilization and lane occupancy rates are increased to different extents, with maximum benefit increments of 17.23% and 18.74%, respectively.