The present dissertation focuses on the development and evaluation of a predictive cooperative driving function, which includes both optimization-based vehicle control and a holistic driving maneuver management. Using realistic traffic simulations, the predictive cooperative driving function is assessed for its potential energy savings and its capability to optimize traffic flow. Initially, a comprehensive simulation environment is established, based on the inner ring of Paderborn’s city center, to generate realistic traffic scenarios. This environment enables an accurate modeling of route planning, traffic light switching programs, and traffic volume. By adjusting the intelligent driver model using real measurement data, the realistic driving behavior of traffic participants is simulated. The parameterization of the simulation scenario is validated based on traffic metrics, ensuring the assessment of the driving strategy under realistic conditions. Two LSTM networks predict the longitudinal velocity of the preceding vehicle and the longitudinal and lateral velocities of neighboring vehicles, especially during a lane change maneuver. The accuracy of these models was validated and overall showed reliable results with minor deviations. A centralized approach is devised for both optimization-based vehicle control and maneuver management. In this approach, the vehicle states of the platoon leader and all following vehicles are simultaneously considered in an optimization problem based on the model predictive control algorithm. The developed driving function allows for an energy-optimized velocity adjustment in both following driving and during various maneuvers. In a microscopic analysis of the simulation scenarios, varying energy-saving potentials are identified based on traffic volume. In uncongested traffic scenarios, it is evident that while the model reduces the distance between vehicles, it occasionally leads to delays and adverse effects for neighboring traffic participants. Hence, the total energy-saving potential of all platoon vehicles is 2.92 %, and the reduction in average velocity is 8.68 %. In congested traffic, the comprehensive coordination of the driving function results in more efficient maneuvers and improved traffic flow. The energy-saving potential ranges between 7.89 % and 9.14 %, and the increase in average velocity is between 3.66 % and 12.06 %. On a macroscopic level, it was determined that the developed function enables energy savings of 0.58 % and 1.62 % for all vehicles in simulations during free-driving and steady traffic conditions. In congested traffic, a marginal increase in energy demand of 0,34 % is observed. In correlation, an increase in the average velocity across all traffic scenarios of 0.47 %, 1.91 %, and 2.94 % is achieved. Furthermore, fundamental diagrams confirm the advantages of the anticipatory and optimization-based velocity control, which offers better adaptation of the velocity to dynamic traffic conditions and contributes to stabilizing the traffic flow.
Prädiktive kooperative Fahrfunktion zur energie- und verkehrseffizienten Fahrzeugführung
Predictive cooperative driving function for energy- and traffic-efficient vehicle control
2024-01-01
1 Online-Ressource : Illustrationen pages
Dissertation, Rheinisch-Westfälische Technische Hochschule Aachen, 2024; Aachen : RWTH Aachen University 1 Online-Ressource : Illustrationen (2024). = Dissertation, Rheinisch-Westfälische Technische Hochschule Aachen, 2024
Miscellaneous
Electronic Resource
German
Prädiktive kooperative Fahrfunktion zur energie- und verkehrseffizienten Fahrzeugführung
TIBKAT | 2024
|Springer Verlag | 2015
|Kooperative Optimierung von Lichtsignalsteuerung und Fahrzeugführung
Tema Archive | 2010
|European Patent Office | 2023
|