Modeling present and future freeway management strategies: variable speed limits, lane-changing and platooning of connected autonomous vehicles
- Autores: Marcel Sala
- Directores de la Tesis: Francesc Soriguera Marti (dir. tes.)
- Lectura: En la Universitat Politècnica de Catalunya (UPC) ( España ) en 2019
- Idioma: español
- Tribunal Calificador de la Tesis: Enrique Castillo Ron (presid.), Francesc Robusté Antón (secret.), Felipe Jiménez Alonso (voc.)
- Programa de doctorado: Programa de Doctorado en Ingeniería Civil por la Universidad Politécnica de Catalunya
- Materias:
- Texto completo no disponible (Saber más ...)
- Resumen
Freeway traffic management is necessary to improve capacity and reduce congestion, especially in metropolitan freeways where the rush period lasts several hours per day. Traffic congestion implies delays and an increase in air pollutant emissions, both with harmful effects to society. Active management strategies imply regulating traffic demand and improving freeway capacity. While both aspects are necessary, the present thesis only addresses the supply side.
Part of the research in traffic flow theory is grounded on empirical data. Today, in order to extend our knowledge on traffic dynamics, detailed and high-quality data is needed. To that end, the thesis presents a pioneering data collection campaign, which was developed in a freeway accessing Barcelona. In a Variable Speed Limits (VSL) environment, different speed limits where posted, in order to observe their real and detailed effects on traffic. All the installed surveillance instruments were set to capture data in the highest possible level of detail, including video recordings, from where to count lane-changing maneuvers. With this objective, a semi-automatic method to reliably count lane changes form video recordings was developed and is presented in the thesis.
Data analysis proved that the speed limit fulfillment was only relevant in sections with enforcement devices. In these sections, it is confirmed that, the lower the speed limit, the higher the occupancy to achieve a given flow. In contrast, the usually assumed mainline metering effect of low speed limits was not relevant. This might be different in case of stretch enforcement. These findings mean that, on the one hand, VSL strategies aiming to restrict the mainline flow on a freeway by using low speed limits will need to be applied carefully, avoiding conditions as the ones presented here. On the other hand, VSL strategies trying to get the most from the increased vehicle storage capacity of freeways under low speed limits might be rather promising.
Results also show that low speed limits increase the speed differences across lanes for moderate demands. This, in turn, also increases the lane changing rates. In contrast, lower speed limits widen the range of flows under uniform lane flow distributions, so that, even for moderate to low demands, the under-utilization of any lane can be avoided. Further analysis of lane-changing activity allowed unveiling that high lane-changing rates prevent achieving the highest flows. This inverse relationship is modeled in the thesis using a stochastic model based on Bayesian inference. This model could be used as a control tool, in order to determine which level of lane-changing activity can be allowed to achieve a desired capacity with some level of reliability.
Previous results identify drivers' fulfillment of traffic regulations as a weak point in order to maximize the benefits of current management strategies, like VSL or lane-changing control. This is likely to change in the near future with the irruption of Autonomous Vehicles (AV) in freeways. V2X communications will allow directly actuating on individual vehicles with high accuracy. This will open the door to new management strategies based on simultaneous communication to groups of AVs and extremely short reaction times, like platooning, which stands out as a strategy with a huge potential to improve freeway traffic. Strings of AVs traveling at extremely short gaps (i.e. platoons) allow achieving higher capacities and lower energy consumption rates. In this context, the thesis presents a parsimonious macroscopic model for AVs platooning in mixed traffic (i.e. platoons of AVs travelling together with human driven vehicles). The model allows determining the average platoon length and reproducing the overall traffic dynamics leading to higher capacities. Results prove that with a 50% penetration rate of AVs in the lane, capacity could reach 3400 veh/h/lane under a cooperative platooning strategy.
