An Optimal Control Method of Intersection with an Upstream Freeway Off-ramp Using Pre-signals

To relieve traffic congestion caused by t h e weaving of traffic s t r e a m s between the expressway off-ramp and the downstream connected intersection, an optimization control method for the intersection with an urban expressway off-ramp was proposed based on pre-signals. The operation process of vehicles from entering to leaving the junction was analyzed. By combining arrival and departure diagrams of vehicles under different conditions, the delay model was established. This model is composed of weaving delay at the ramp junction, pre-signal control delay, and main signal control delay. Consequently, a traffic organization optimization model based on prc-signal for off-ramp junction intersection of expressway was established. The objective of this model is to minimize the average vehicle delay at the junction and intersection. The constraints of this model include intersection signal timing, lane allocation, and queue length. These issues were solved using the particle swarm optimization algorithm. Typical expressway and junction intersections in Changchun were used to verify the developed model. Six traffic scenarios were designed to verify the improvement of traffic benefits under different geometric conditions, left turn traffic proportions, and intersection saturation. The results show that by setting pre-signals and sorting areas, the vehicle delay and stop times at the intersection are significantly reduced by 27. 94% and 12. 00%, respectively, and the off-ramp traffic flow is reduced the most. The sensitivity analysis shows that the pre-signal control model proposed in this study can be applied to different traffic scenarios. Additionally, greater traffic benefits can be achieved when the intersection has more lanes and a large proportion of left-turn traffic. The research results can provide support for the traffic control in the connection area of urban expressway off-ramp. © 2023 Xi'an Highway University. All rights reserved.