Abstract-During peak hours in urban areas, unpredictable traffic congestion caused by en route events (e.g., vehicle crashes) increases drivers' travel time and, more seriously, decreases their travel time reliability. In this paper, an original and highly practical vehicle rerouting system, which is called Next Road Rerouting (NRR), is proposed to aid drivers in making the most appropriate next road choice to avoid unexpected congestions. In particular, this heuristic rerouting decision is made upon a cost function that takes into account the driver's destination and local traffic conditions. In addition, the newly designed multiagent system architecture of NRR allows the positive rerouting impacts on local traffic to be disseminated to a larger area through the natural traffic flow propagation within connected local areas. The simulation results based on both synthetic and realistic urban scenarios demonstrate that, compared with the existing solutions, NRR can achieve a lower average travel time while guaranteeing a higher travel time reliability in the face of unexpected congestion. The impacts of NRR on the travel time of both rerouted and nonrerouted vehicles are also assessed, and the corresponding results reveal its higher practicability.

A hybrid automaton consists of a discrete state component represented by a finite automaton, coupled with a (vector) continuous state component governed by a differential equation. For hybrid automata it is possible to reduce certain verification problems to those of checking language containment or language emptiness. Here we present a class of hybrid automata called suspension automata for which conditions can be given under which these problems are decidable.

As urbanization has been spreading across the world for decades, the traffic congestion problem becomes increasingly serious in most of the major cities. Among the root causes of urban traffic congestion, en route events are the main source of the sudden increase of the road traffic load, especially during peak hours. The current solutions, such as on-board navigation systems for individual vehicles, can only provide optimal routes using current traffic data without considering any traffic changes in the future. Those solutions are thus unable to provide a better alternative route quickly enough if an unexpected congestion occurs. Moreover, using the same alternative routes may lead to new bottlenecks that cannot be avoided. Thus a global traffic load balance cannot be achieved. To deal with these problems, we propose a Multi Agent System (MAS) that can achieve a trade-off between the individual and global benefits by giving the vehicles optimal turn suggestions to bypass a blocked road ahead. The simulation results show that our strategy achieves a substantial gain in average trip time reduction under realistic scenarios. Moreover, the negative impact of selfish re-routing is investigated to show the importance of altruistic re-routing applied in our strategy.

Some rights reserved. For more information, please see the item record link above.

Abstract-Due to the severe impact of road traffic congestion on both economy and environment, several vehicles routing algorithms have been proposed to optimize travelers itinerary based on real-time traffic feeds or historical data. However, their evaluation methodologies are not as compelling as their key design idea because none of them had been tested under both real transportation map and real traffic data. In this paper, we conduct a deep performance analysis and comparison of four typical vehicles routing algorithms under various scalability levels (i.e. trip length and traffic load) based on realistic transportation simulation. The ultimate goal of this work is to suggest the most suitable routing algorithm to use in different transportation scenarios, so that it can provide a valuable reference for both traffic managers and researchers when they deploy or optimize a large scale centralized Traffic Management System (TMS). The obtained simulation results reveal that dynamic A* is the best routing algorithm if the TMS has sufficient memory or storage capacities, otherwise static A* is also a great alternative.