Bernhard Friedrich

A decentralized adaptive traffic signal control using V2I communication data

A novel concept for a decentralized adaptive traffic signal control in urban networks using in future available vehicle to infratructure (V2I) communication data is presented. The phase-based strategy takes advantage of the improved detection data and optimizes each time interval of 5 seconds the phase sequence in order to reduce the total queue length within a forecast horizon of 20 seconds. For optimization the methods of dynamic programming and complete enumeration are used. The methods are embedded in the simulation environment of the microscopic traffic simulator AIMSUN NG. The market penetration level is the critical factor that impacts the quality of the new signal control. Hence, various penetration levels are modelled. For reference TRANSYT-7F is used.

Exploitability of vehicle related sensor data in cooperative systems

This paper shows how extended vehicle related sensor data is gained and used within the cooperative system developed in the project KOLINE. This system aims at coordinating traffic in order to achieve a reduction of delay and emissions in urban networks by exchanging information between traffic signals and approaching vehicles. Besides receiving information about the signal program and traffic state at the intersection the vehicle also uses data of multiple sensors and different sensor technologies. This allows the creation of a redundant representation of the vehicles environment which is used to improve the vehicles approach to intersections. Additionally, the information can be used by the infrastructure to improve the performance of the adaptive traffic control system by using it for tailback approximation and for the calibration of the underlying traffic model.

Online control of signalized networks using the Cell Transmission Model

Adaptive Traffic Control Systems (ATCS) aim at the improvement of traffic flow in terms of delay, number of stops, travel time etc. in urban networks under the regime of traffic signal control. Well known ATCS are SCOOT, SCATS, BALANCE and MOTION to name only a few. Lately the Cell Transmission Model (CTM) attracted some attention in the context of traffic signal control. It can be used to model the effects of different signal settings. Based on the CTM a prototype of a new ATCS has been developed. It can handle urban sub-networks with several interconnected signalized intersections. Every 15 minutes it adapts and optimizes signal plans and coordination patterns to the currently estimated traffic demand in the network. In a first step the upcoming traffic demand is forecasted and estimated. Based on this demand cycle length and green splits are adjusted. Finally, offsets are optimized by using the CTM in combination with Genetic Algorithms and a second alternative approach named Sequential Enumeration. The emphasis of this paper is on the offset optimization and the performance of the new ATCS which has been investigated in a microsimulation study.

Data Fusion Techniques for Adaptive Traffic Signal Control

Abstract Adaptive online traffic control in urban road networks requires information on the present and future traffic state. This information needs to be as complete and precise as possible. This paper introduces on the one hand a new method for online queue length determination based on data fusion technique (Mueck, 2002) and investigates on the other hand the performance gains that can be achieved employing this determination module as a quasi measurement with Kalman filtering technique in queuing theory models (e. g. Markovian chains). Both, the theoretical approach of the respective models as well as the results of simulation studies are presented.

Verkehrliche Wirkung autonomer Fahrzeuge

Autonome Fahrzeuge nehmen selbststandig am Verkehr teil, ohne dass sie den Menschen als Uberwacher oder Entscheider benotigen. Ihren Fahrgasten bieten autonome Fahrzeuge einen Komfortgewinn, da keine Fahraufgaben geleistet werden mussen. Einem Personenkreis, der bislang aufgrund von Mobilitatseinschrankungen von der Teilhabe am offentlichen Leben teilweise oder ganz ausgeschlossen ist, bieten autonome Fahrzeuge neue Chancen fur dessen Mobilitat.

The Effect of Autonomous Vehicles on Traffic

Autonomous vehicles maneuver in traffic through road networks without requiring humans as supervisors or decision makers. Autonomous vehicles increase comfort for their passengers by removing the need for them to perform driving tasks. Autonomous vehicles provide new mobility opportunities for groups of people that thus far have been partially or entirely excluded from participation in public life due to mobility restrictions.

Traffic Signal Transition in Coordinated Meshed Networks

Transition of signal plans in networks with controlled traffic lights always has a disruptive impact on traffic flow. The extent of these disruptions depends on the transition method, and therefore increases in delay or travel time may be more or less severe. The task of transitioning becomes even more complex in coordinated networks in which new coordination patterns have to be established quickly without inducing major disturbances. Several transition methods exist, some of which are rather abrupt; others are smoother. Many of these methods have been analyzed in simulation studies. Two extensive studies investigated not only transition effects at isolated intersections but also transition effects on coordinated arterials. This paper adds to these studies by investigating transition effects in a meshed network with different coordinated relations. Furthermore, an additional transition method and variations of previously tested methods have been considered. Four scenarios have been set up and simulated. The results of this study are discussed in detail, delivering not only information on the overall performance of each method but also allowing for some further insights into what is happening during transition. Some effects at intersections are induced by transition at that very intersection, while others have to be ascribed to neighboring intersections. Dwell methods are unfavorable for undisturbed transitioning; smooth methods perform better, especially if transition takes no more than three cycles. The results of this study are valuable for both time-of-day signal plan selection and for adaptive traffic control strategies.

A Multiagent Approach to Modeling Autonomic Road Transport Support Systems

In this chapter, we investigate a multiagent based approach to modeling autonomic features in urban traffic management. We provide a conceptual model of a traffic system comprising traffic participants modeled as locally autonomous agents, which act to optimize their operational and tactical decisions (e.g., route choice), and traffic management center(s) (TMC) which influence the traffic system according to dynamically selected policies. In this chapter, we focus on two autonomic features which emerge from the local decisions and actions of traffic participants and their interaction with the TMC and other vehicles: (1) Autonomic routing, in which we study how vehicle agents can individually adapt routing decisions based on local learning capabilities and traffic information communicated truthfully by a traffic management center; and (2) Autonomic grouping, i.e., collective decision-making of vehicles, which exchange route information and dynamically form and operate groups to drive in a convoy, thus aiming at higher speed and increased throughput. Communication is based on a (simulated) vehicle-to-infrastructure (V2I) and vehicle-to-vehicle (V2V) protocols. Initial experiments are reported using a real-world traffic scenario modeled in the Aimsun software, which is extended by the decision logic of TMC and vehicles. The experiments indicate that autonomic routing and grouping can improve the performance of a traffic management network, even though negative effects such as unstable behavior can be observed in some cases.

Concurrent Split, Offset and Cycle Control through Model Predictive Control

Abstract The paper presents an extension to an online network-wide urban traffic control technique, that uses the Model Predictive Control (MPC) framework. The extension adds offset and cycle equations to the existing systems model. The resulting control scheme takes into consideration desired cycle and offset values when calculating greens durations. This new feature creates a balancing effect that weights in queue management and offset tunning, apart from allowing the manipulation of offsets between secondary stages. A comparison between the proposed extension, the original solution, and a TRANSYT optimized fixed-time control shows promising results.

A MODULAR SYSTEM FOR TRAFFIC FLOW SIMULATIONS AT INTERSECTIONS AND IN NETWORKS

Traffic planning and traffic control necessitate a tool to evaluate the performance of the considered traffic system. This requires a multi-purpose calculation method allowing to compute delays, queue lengths etc. for different planning alternatives. These effects can be obtained from the modular system SIMPL for traffic flow simulation described in this paper. Formulating the model components with respect to recent research studies, a representation of traffic flow close to reality can be achieved. On the basis of a microscopic, time and space discrete concept, the simulation model gives the opportunity to integrate additional parameters considering real road user behaviour. In order to avoid the “black box effect” a graphical real time presentation of the simulated traffic provides a valuable tool to detect program inconsistencies and to assess verisimilitude of the model.