Micha Koller

Microscopic Simulation of Synchronized Flow in Oversaturated City Traffic: Effect of Driver’s Speed Adaptation

Synchronized flow has been found recently in studies of empirical GPS probe-vehicle data collected in oversaturated city traffic. In this paper, results of simulations with a three-phase microscopic model from Kerner and Klenov are presented. These results show features of synchronized flow. The effect of the drivers speed adaptation was found to play the key role in understanding the emergence of synchronized flow in over-saturated city traffic. The physical meaning of the speed adaptation effect in oversaturated city traffic was explained, and the influence of the speed adaptation effect on the average speed and travel time in oversaturated city traffic was investigated.

Increased Consumption in Oversaturated City Traffic Based on Empirical Vehicle Data

Congestion of urban roads causes extra travel time as well as additional fuel consumption. We present an approach to determine this additional fuel consumption on the basis of empirical vehicle data. We study probe vehicle data provided by TomTom to find the various traffic patterns of urban congestion. We use simulations of these urban traffic patterns based on a stochastic Kerner- Klenov model as input for an empirical fuel consumption matrix compiled from empirical CAN bus signals from vehicles. Our results confirm that in certain congested city traffic patterns vehicles consume more than twice as much fuel as in free city traffic.

Spatio-Temporal Traffic Pattern Recognition Based on Probe Vehicles

Ubiquitous mobile probe data give new opportunities for the precise reconstruction of congested traffic situations. Kerner’s three-phase traffic theory (Kerner, The physics of traffic. Springer, Berlin/Heidelberg/New York, 2004; Introduction to modern traffic flow theory and control. Springer, Berlin/Heidelberg, 2009) is the theoretical fundament for a review of the probe data analysis presented in this paper. The methodology of the approach developed initially in Kerner et al. (Physica A 392:221–251, 2013), will be illustrated and evaluated with empirical examples from a German field trial. The mobile probes are processed with a three-phase traffic state recognition while the vehicles drive through a spatio-temporal congested traffic pattern, i.e., they pass synchronized flow regions and/or wide moving jams. In the traffic control center, the traffic states from all communicating mobile probes are fused depending on the related traffic phase. The quality of the reconstructed traffic pattern using the mobile probes can be correlated with the reconstruction based on stationary detectors. Therefore, we can conclude which amount of mobile probes give the same information accuracy as roadside detectors at certain distances. A microscopic traffic simulation based on Kerner-Klenov traffic model (Kerner and Klenov, J Phys A Math Gen 35:L31–L43, 2002; Phys Rev E 68:036130, 2003; J Phys A Math Gen 37:8753–8788, 2004; Phys Rev E 80:056101, 2009) has given us an environment for developing, testing and evaluation of the traffic reconstruction algorithms. The car-to-infrastructure field trial with more than 120 vehicles communicating with the traffic control center for the duration of 6 month in the German federal state of Hessen produces a huge amount of empirical data. The paper illustrates results of the congested traffic recognition and jam front detection. We will show that 2 % communicating probe vehicles of the total flow rate give the opportunity of precise jam front warnings and, in addition, the same data quality as detectors of 1–2 km distances.

Traffic Phase Dependent Fuel Consumption

Fuel consumption is one of the key cost factors relevant for the movement of vehicles. In times of increasing traffic congestion on both freeways and urban road sections the question arises how the fuel consumption is influenced by congestions congestion occurring in many sections of the road network. Congested traffic states are defined based on Kerner’s three-phase traffic theory [1, 2]. The article presents the probability functions of traffic breakdowns for road sections: the probability curve as function of the traffic flow rate is an increasing function of the flow rate and similar for both freeway an urban sections with traffic signals [4]. Therefore, the recognition of traffic breakdowns and the determination of the emerging traffic state is crucial for the prediction of the additional fuel consumption. By investigating empirical field data from vehicles driving on a specific freeway section statistical analysis reveals the additional fuel consumption factors for the two different congested states in comparison to free flow.

Simulations of Synchronized Flow in TomTom Vehicle Data in Urban Traffic with the Kerner-Klenov Model in the Framework of the Three-Phase Traffic Theory

In this article, we describe our simulations of TomTom probe vehicle data measured in city traffic. An analysis of the vehicle trajectories in the TomTom data reveals the typical features of the traffic phases as defined in Kerner’s three-phase traffic theory: free flow, synchronized flow and wide moving jam (moving queues). The existence of the synchronized flow phase has previously been found within traffic data from highways, but not within data from urban road networks. We will show that the microscopic simulation of vehicular traffic with the stochastic Kerner-Klenov model on a multi-lane urban road stretch reproduces the synchronized flow found in the TomTom data.