Computationally efficient model predictive control of freeway networks

Evaluating Performance of a Proactive Optimal Variable Speed Limit Control Using Different Objective Functions

Variable Speed Limit (VSL) control contributes to potential crash risk reduction by suggesting a suitable dynamic speed limit to achieve more stable and uniform traffic flow. In recent studies, researchers adopted macroscopic traffic flow models and perform prediction-based optimal VSL control. The response of drivers to the advised VSL is one of the most critical parameters in VSL-controlled speed dynamics modeling, which significantly affects the accuracy of traffic state prediction as well as the control reliability and performance. Nevertheless, the variations of driver responses were not explicitly modeled. Thus, in this research, the authors proposed a dynamic driver response model to formulate how the drivers respond to the advised VSL during various traffic conditions. The model was established and calibrated using field data to quantitatively analyze the dynamics of drivers’ desired speed regarding the advised VSL and current traffic state variables. A proactive VSL control algorithm incorporating the established driver response model was designed and implemented in field-data-based simulation study. The design proactive control algorithm modifies VSL in real-time according to the traffic state prediction results, aiming to reduce potential crash risks over the experiment site. By taking into account the real-time driver response variations, the VSL-controlled traffic state dynamics was more accurately predicted. The experimental results illustrated that the proposed control algorithm effectively reduces the crash probabilities in the traffic network.

Author(s): Gao, Hang; Cheng, Shenyang; Zhang, Michael | Abstract: The model based variable speed limit (VSL) control has been proven effective to resolve capacity-drop and time delay at a single recurrent bottleneck in previous studies. This project applies VSL controls to the traffic corridors with multi-segment and multi-bottleneck with the objective of reducing fuel consumption and greenhouse gas emissions. Based on a comprehensive review of existing methods, we develop and compare two fuel consumption centered VSL control (FC-VSL) strategies: flow-based control versus density-based control. These control strategies are implemented in SUMO, a microscopic traffic simulation package, on a 10-mile long freeway section. Results show that the density-based control reduces fuel consumption and gas emissions significantly at the cost of slight increase of travel time. The flow-based control, in contrast, reduces congestion and emissions in the downstream segments but transfers the congestion to the segments upstream of the controlled segments, resulting in an overall performance that is worse than the density-based FC-VSL, and no better than imposing static speed limits.

The prevailing variable speed limit (VSL) systems as an effective strategy for traffic control on motorways have the disadvantage that they only work with static VSL zones. Under changing traffic conditions, VSL systems with static VSL zones may perform suboptimally. Therefore, the adaptive design of VSL zones is required in traffic scenarios where congestion characteristics vary widely over space and time. To address this problem, we propose a novel distributed spatial-temporal multi-agent VSL (DWL-ST-VSL) approach capable of dynamically adjusting the length and position of VSL zones to complement the adjustment of speed limits in current VSL control systems. To model DWL-ST-VSL, distributed W-learning (DWL), a reinforcement learning (RL)-based algorithm for collaborative agent-based self-optimization toward multiple policies, is used. Each agent uses RL to learn local policies, thereby maximizing travel speed and eliminating congestion. In addition to local policies, through the concept of remote policies, agents learn how their actions affect their immediate neighbours and which policy or action is preferred in a given situation. To assess the impact of deploying additional agents in the control loop and the different cooperation levels on the control process, DWL-ST-VSL is evaluated in a four-agent configuration (DWL4-ST-VSL). This evaluation is done via SUMO microscopic simulations using collaborative agents controlling four segments upstream of the congestion in traffic scenarios with medium and high traffic loads. DWL also allows for heterogeneity in agents’ policies; cooperating agents in DWL4-ST-VSL implement two speed limit sets with different granularity. DWL4-ST-VSL outperforms all baselines (W-learning-based VSL and simple proportional speed control), which use static VSL zones. Finally, our experiments yield insights into the new concept of VSL control. This may trigger further research on using advanced learning-based technology to design a new generation of adaptive traffic control systems to meet the requirements of operating in a nonstationary environment and at the leading edge of emerging connected and autonomous vehicles in general.

Exploring the impact of a coordinated variable speed limit control on congestion distribution in freeway

Ahstract- Different types of variable speed limit (VSL) controllers have been proposed to improve the traffic mobility at highway bottlenecks. However, the benefit of VSL is sometimes deteriorated by the capacity drop phenomenon introduced by the disordered lane changes at the vicinity of the bottleneck. Our recent results show that by combining the VSL controller with lane change control, the capacity drop can be significantly relieved or avoided at the bottleneck. In this paper, we propose feedback linearization and model predictive control schemes for VSL-actuated highway traffic, where a lane change controller is assumed to be active upstream of the bottleneck. Via simulation studies, we compare the performance and robustness of the proposed controllers with respect to perturbations on traffic demand, model parameters and measurement noise. Results show that both controllers are able to improve the total time spent under different levels of perturbation and noise. Furthermore, feedback linearization VSL can provide better performance than model predictive VSL with negligible computational effort and similar robustness.

Freeway corridor traffic flow is limited by bottleneck flow. If the section upstream of a bottleneck is congested, the bottleneck flow will drop well below its capacity. A logical approach to maximizing recurrent bottleneck flow is to create a discharge section immediately upstream of the bottleneck. This paper proposes a control strategy for combining variable speed limits (VSL) and coordinated ramp metering (CRM) design to achieve this objective when the bottleneck can be represented as a lane (or virtual lane) reduction. At each time step, VSL is designed first, with mainline flow, on-ramp demand and length, and driver acceptance taken into account. VSL values are determined for three locations: (a) the critical VSL to regulate the discharge section flow to match bottleneck capacity, (b) the VSL for the potentially congested section (mainline storage), and (c) the VSL upstream of the congestion tail. With the selected VSL, the first-order mainline flow model is linearized and then used for CRM design with model predictive control to minimize the difference between scaled total travel time (or vehicle hours traveled) and the total travel distance (equivalent to vehicle miles traveled). Microscopic simulation shows that this control strategy should improve bottleneck throughput significantly.

Dynamic areas strategy design for variable speed limiting at fixed freeway bottlenecks

In recent years, congestion on port motorways has become increasingly frequent, significantly constraining transportation efficiency and contributing to higher pollution emissions. This paper proposes a novel max-pressure-driven integrated control (IFC-MP) for port motorways, inspired by the max pressure (MP) concept, which continuously adjusts the weights of ramp metering (RM) and the variable speed limit (VSL) based on pressure feedback from the on-ramp and upstream, assigning greater control weight to the side with higher pressure. A queue management mechanism is incorporated to prevent on-ramp overflow. The effectiveness of IFC-MP is verified in SUMO, filling the gap where the previous integrated control methods for port motorways lacked micro-simulation validation. The results show that IFC-MP enhances bottleneck throughput by approximately 7% compared to the no-control case, optimizes the total time spent (TTS) by 26–27%, and improves total pollutant emissions (TPEs) by about 11%. Compared to strategies that use only RM and VSL control, or activate VSL control only after RM reaches its lower bound, the time–space distribution of speed under IFC-MP is more uniform, with smaller fluctuations in bottleneck occupancy. Additionally, IFC-MP maintains relatively stable performance under varying compliance levels. Overall, the IFC-MP is an effective method for alleviating congestion on port motorways, excelling in optimizing both traffic efficiency and pollutant emissions.

The variable speed limit (VSL) is a control instrument of emerging intelligent transportation systems used to help manage motorway traffic. The VSL has the potential to improve traffic safety and reduce both traffic congestion and emissions through better harmonisation of traffic flow. The VSL results in a reduction of speed variations among vehicles travelling along controlled sections. In this paper, a modified logic-tree-based algorithm is proposed to improve the mobility benefits of VSL controllers. The modified algorithm is assessed using a case study in Auckland, New Zealand. Analysis of the case study is based on a critical bottleneck section of the Auckland motorway using the Aimsun micro-simulator. The proposed algorithm is compared with a known algorithm proposed in the literature and used in practice. The simulation results of the case study show that the proposed logic-tree-based VSL control algorithm outperforms the existing method. Compared with a no-control scenario, it results in respective improvements of 6%, 11%, 3%, 9% and 3% in the capacity of the bottleneck section, the total travel time and emissions of carbon dioxide, carbon monoxide and nitrogen oxides.

A deep reinforcement learning-based intelligent intervention framework for real-time proactive road safety management

Congestion has become highly recognized as a worldwide traffic problem, as traffic demand has grown steadily over the past few decades. Variable speed limits (VSLs) are an intelligent transportation system (ITS) measure that limits mainline flow to mitigate bottleneck congestion. Currently, VSLs have become proactive based on short-term prediction. Proactive VSLs succeed in simulation evaluations, but few have been deployed in the field and their real-world effectiveness has not been proven. Various factors may lead to this limitation, such as the absence of reliable field application software, accuracy of prediction models, and high computation time for proactive control. To address this research gap, this study reports a preliminary VSL test and details its implementation results on Whitemud Drive, Edmonton, Canada. First, based on field traffic measurements before VSL control, recurrent bottleneck locations are identified. Second, the proactive control algorithm is briefly introduced. Then, a s...

RaceCC: A rapidly converging explicit congestion control for datacenter networks

Despite many years of improvements to it, TCP still suffers from an unsatisfactory performance. For services dominated by short flows (e.g., web search and e-commerce), TCP suffers from the flow startup problem and cannot fully utilize the available bandwidth in the modern Internet: TCP starts from a conservative and static initial window (IW, 2-4 or 10), while most of the web flows are too short to converge to the best sending rate before the session ends. For services dominated by long flows (e.g., video streaming and file downloading), the congestion control (CC) scheme manually and statically configured might not offer the best performance for the latest network conditions. To address these two challenges, we propose TCP-RL, which uses reinforcement learning (RL) techniques to dynamically configure IW and CC in order to improve the performance of TCP flow transmission. Basing on the latest network conditions observed at the server side of a web service, TCP-RL dynamically configures a suitable IW for short flows through group-based RL, and dynamically configures a suitable CC scheme for long flows through deep RL. Our extensive experiments show that for short flows, TCP-RL can reduce the average transmission time by about 23%; and for long flows, compared with the performance of 14 CC schemes, TCP-RL's performance ranks top 5 for about 85% of the 288 given static network conditions, whereas for about 90% of conditions, its performance drops by less than 12% compared with that of the best-performing CC schemes for the same network conditions.

Robust variable speed limit control with respect to uncertainties