Effect of Proactive Braking on Traffic Flow and Road Throughput

Abstract

Abstract-With the emergence of pervasive computing technologies into vehicles, driving has moved from an active task of steering towards an interaction or adaptation task with respect to the driver-vehicle feedback loop. Up to now vehicular interfaces have mostly been evaluated from a single-driver singlecar viewpoint, however, driving is a more complex task involving -beside the local interaction -the interrelationship between all the cars in a certain community of interest.The question investigated in this research work is how a vehicle's local parameters in a bulk of cars (e. g. vehicle speed, braking parameters) affect the global behavior of this system (traffic congestion, driving speed variation, throughput). To explore this, two traffic models have been developed and simulated using the NetLogo simulation environment.Simulation results have shown that the intercar distance has a direct impact on both the throughput and the mean trip time. The proactive driving approach using vibro-tactile driver notification followed in the second, advanced model achieved much better results regarding these parameters compared to the simple manual-driven case. Finally, the outcomes legitimate the implementation of a prototype, and the installation of such a technology into a large number of cars in order to provide evidence for the improved traffic flow and decreased probability of traffic accidents in real driving scenarios.Index Terms-Traffic congestion, Stop-and-go traffic, Trip time, Intercar distance, Throughput, ACC, Vibro-tactile notification, Complex adaptive systems. I. PERVASIVE COMPUTING REVOLUTIONIZES TRAFFICSince the diffusion of pervasive computing technologies into cars both vehicle handling (steering, braking, accelerating, etc.) and driving comfort (route guiding, communication, driving assistance, entertainment, etc.) have changed massively [1] and have led, particularly for the formerly traditional controls, to a strong interrelationship between the driver and the technical systems in a car.Up to now, most of these advances concern the interaction between a single driver and a single car. With latest achievements in wireless communication technologies, a new class of vehicle-to-"x" applications arises, allowing the spontaneous formation of collections of cars (cooperative crowds) to offer car-to-car (safety functions, proactive traffic jam avoidance, accident prevention, negotiation of driving parameters, etc.), car-to-roadside and car-to-infrastructure applications. A. Complex Adaptive Systems (CASs)Bulks of cars on a road segment of interest can be assumed to be complex adaptive systems, as it can be observed that every driver-car pair acts on its own interests (following its local navigation goal using different strategies such as reaching the destination as fast as possible, as cheap as possible with respect to road toll, as short as possible regarding the distance, etc.) and affected by the personal style of driving, leading to a (apparently) random behavior of the collection of these pairs. Howeve...

OnTraJaCS: Ontology based Traffic Jam Control System

“…AndreasRiener and AloisFerscha investigated the effect of individual vehicle parameters on the behavior of the whole group in a specific road segment [9]. The team's simulations have shown that the greatest factor affecting throughput and mean trip time is inter-car distance.…”

An urban traffic simulation system based on multi-agent modeling

“…It shows that the agent-based technology is helpful to analyze the effect of different light plans on average for the ITS. In [4], the agent-based technology is used in studying the effect of proactive braking on traffic flow and road throughput in the traffic system. All these work in [3,4,5] shows that the agent-based traffic simulation is greatly helpful to study the dynamic traffic system's behavior and show the autonomy and interaction of traffic elements better than the conventional traffic models.…”