Abstract: In order to ease the congestion problem in urban intersections and overcome the drawbacks of the tradi- tional traffic-responsive strategy,a real-time vehicle queue length estimation algorithm for the intersections in the environment of partial internet-of-vehicles is proposed,and a traffic flow priority calculation model is constructed to minimize delay time.On this basis,by comprehensively considering such factors as the traffic safety at intersec- tions,an adaptive intersection control method with dynamic traffic flow arrangement is designed.In order to verify the accuracy and feasibility of the proposed model and method,the above-mentioned control logic is implemented by the VISSIM-COM programming,and a typical intersection is selected to conduct a simulation.The result shows that (1) the proposed queue length estimation algorithm is of a higher accuracy; (2) as compared with the tradi- tional traffic-responsive strategy and the fixed-time strategy,the proposed control model achieves an 70% increase in average queue length，with an average delay decrease of about 65% and 55% respectively; and (3) the pro- posed model makes up for the deficiency of the traditional traffic-responsive strategy in easing the traffic near or over the saturation point,and effectively improves the operating efficiency at intersections.

Key words: urban traffic, connected vehicle network, intersection, queue length estimation, adaptive control