Autonomous vehicles create new opportunities for innovative intelligent traffic systems. Variable speed limit (VSL), which is a speed management system that can adjust the speed limit according to traffic conditions or a predefined speed control algorithm on different road segments, can be better implemented with the cooperation of autonomous vehicles. These compliant vehicles can automatically follow speed limits. However, noncompliant vehicles will attempt to pass the moving bottleneck created by compliant vehicles. This study builds a multiclass cell-transmission model (CTM) to represent the relation among traffic flow parameters. This model calculates flows of both compliant and noncompliant vehicles between road segments. An VSL algorithm is proposed to reduce the stop-and-go behavior of vehicles at traffic signals and to save energy consumption. Simulation is used to test the effects of VSLs on an example network. The result shows that VSLs are effective at reducing the energy consumption of the whole system.
8 June 2022 in Transportation Research Record: Journal of the Transportation Research Board
Max-pressure control is a decentralized method of traffic intersection control, making computations at individual intersections simple. In addition, this method of control has been proven to maximize network throughput if any traffic signal control can stabilize the demand. This paper tests max-pressure controllers in a large-scale microsimulation of the downtown Austin network using the microscopic traffic simulation package SUMO. Nine combinations of weight function and method of defining green time are studied to see how different variations on the max-pressure controller compare. It is shown that the way green time is assigned (cyclic or non-cyclic) has a larger impact on performance than the weight function used by the max-pressure controller. Based on these results a new way of assigning green time is devised. This novel controller mirrors the performance of either the cyclic or the non-cyclic controller depending on the geometry and demand. Large-scale simulation shows that this controller compares favorably with existing controllers using metrics of number of waiting vehicles and average travel time. Common problems with non-cyclic control include the higher likelihood of gridlock and the potential for very long waiting times when demand at a single intersection is asymmetric. On the other hand, the cyclic controller is required to allocate green time to every phase even if the demand is low, increasing the loss time. The novel semi-cyclic controller solves these inherent problems with the cyclic and non-cyclic controllers, making it more likely to be implemented by traffic engineers.
24 April 2024 in IEEE Transactions on Intelligent Transportation Systems
Recently, the limited capacity of the Urban Rail Transit (URT) has failed to meet passenger travel demands, especially in peak hours, which leads to crowded stations and oversaturated metro networks. Considering the diverse scenarios of the multimodal public transportation network and the need for rapid generation of strategies, this study proposes a deep reinforcement learning (RL) algorithm that guides passenger route selection in the multimodal transit network to operate better. The proposed RL algorithm, consisting of graph representation learning, convolution neural network, and self-attention mechanism, is used to generate the route guidance strategy for passengers, aiming to alleviate the congestion of the multimodal transit network, improve passengers’ travel experience, and reduce CO $_{2}$ emissions. Based on the multimodal transit network in Beijing, the simulation results demonstrate that the RL algorithm can well perceive the states and generate adaptive route guidance strategies that can decrease the section load rates, improve network-wide passengers’ travel experience, and reduce CO $_{2}$ emissions. Even in cases where passengers are not fully compliant with the route guidance, the proposed algorithm remains effective.
15 July 2025 in Transportation Research Part C: Emerging Technologies