Characterizing Heterogeneous Traffic Flow at a Slope Bottleneck via Cellular Automaton Model

Abstract

This paper considers the unique attributes of autonomous vehicles, the characteristics of expressways’ minor actual slope grade, and the speed update integer requirements of discrete cellular automata (CA). The continuous deceleration efficiency of the uphill is discretized by introducing variable parameters connected to the vehicle position and gradient, and the hybrid traffic flow CA model suitable for slope bottlenecks based on the principle of the safety distance is further improved. The simulation results show that the improved model can simulate the unique traffic flow phenomenon caused by the change in slope conditions. There is a critical mutation in the traffic flow phase on the slope road section. Secondly, according to the different main factors influencing traffic flow, free flow, terrain jams, and homogeneous congested flow coexist on the slope section. The traffic flow in the crowded terrain area is significantly negatively correlated with the length and longitudinal gradient of the slope, and increasing the longitudinal angle of the hill will extend the congested terrain area. In addition, the influence of the change in slope conditions on mixed traffic flow has a critical longitudinal gradient of 3% and a fixed critical slope length. In non-free flow areas, autonomous vehicles can significantly alleviate traffic congestion on upslopes.