Traffic and Travel Information (TTI) systems in use today are based on a centralized structure. Sensors along the roadside monitor traffic density and transmit the results to a central unit where a situation analysis of the global traffic scenario is performed. The resulting traffic information is made available to the drivers via broadcast service or alternatively on demand via cellular phone.

Cellular automaton (CA) based models are increasingly used to investigate cities and urban systems. We discuss difficulties with this representation of human systems, and suggest that many modifications to simple CA introduced in modelling cities are responses to these problems. We propose a two-pronged approach to research. First, for operational model-building many variations on the CA theme are required and should be welcomed; and second theoretically motivated variations of the CA formalism are required so that the possible effects on model dynamic behaviour may be more systematically explored.

Knowledge of fundamental traffic flow characteristics of traffic simulation models is an essential requirement when using these models for the planning, design, and operation of transportation systems. In this paper we discuss the following: a description of how features relevant to traffic flow are currently under implementation in the TRANSIMS microsimulation, a proposition for standardized traffic flow tests for traffic simulation models, and the results of these tests for two different versions of the TRANSIMS microsimulation.

This paper presents a day-to-day re-routing relaxation approach for traffic simulations. Starting from an initial planset for the routes, the route-based microsimulation is executed. The result of the microsimulation is fed into a rerouter, which re-routes a certain percentage of all trips. This approach makes the traffic patterns in the microsimulation much more reasonable. Further, it is shown that the method described in this paper can lead to strong oscillations in the solutions. I.

Abstract not found

A simple, rule-based approach to traffic flow can yield astonishingly realistic results and is therefore a candidate for very fast large scale microscopic traffic simulations. In the present article, we evaluate two conceptually different codings of the same dynamics on parallel supercomputers. We use a Parsytec GCel-3 (1024 nodes), an Intel iPSC/860 (32 nodes), and a Connection Machine CM-5 (32 nodes). For comparison purposes, we use as well a NEC-SX3/11 traditional single node vector computer, and a net of coupled workstations. Compared to published computing speeds of microscopic traffic models, our model proves to be up to about 1000 times faster. We find our highest computing speed by employing a single-bit coding scheme used in, e.g., Ising-model programming. As traffic flow is a one-dimensional problem, a complication is that geometric parallelization has to be done in the same direction as single-bit coding. Nevertheless, we reach efficiencies near 100 percent for large syst...

We discuss the cellular automata approach and its extensions, the lattice Boltzmann and multiparticle methods. The potential of these techniques is demonstrated in the case of modeling complex systems. In particular, we consider applications taken from various fields of physics, such as reaction-diffusion systems, pattern formation phenomena, fluid flows, fracture processes and road traffic models.

One of the main questions concerning learning in Multi-Agent Systems is: "(How) can agents benefit from mutual interaction during the learning process?". This paper describes the study of an interactive advice-exchange mechanism as a possible way to improve agents' learning performance. The advice-exchange technique, discussed here, uses supervised learning (backpropagation), where reinforcement is not directly coming from the environment but is based on advice given by peers with better performance score (higher confidence), to enhance the performance of a heterogeneous group of Learning Agents (LAs). The LAs are facing similar problems, in an environment where only reinforcement information is available. Each LA applies a different, well known, learning technique: Random Walk, Simulated Annealing, Evolutionary Algorithms and Q-Learning. The problem used for evaluation is a simplified traffic-control simulation. In the following text the reader can find a description of the traffic simulation and Learning Agents (focused on the advice-exchange mechanism), a discussion of the first results obtained and suggested techniques to overcome the problems that have been observed. Initial results indicate that advice-exchange can improve learning speed, although "bad advice" and/or blind reliance can disturb the learning performance. The use of supervised learning to incorporate advice given from non-expert peers using different learning algorithms, in problems where no supervision information is available, is, to the best of the authors' knowledge, a new concept in the area of Multi-Agent Systems Learning.

This paper describes the first impressions of the development of a multi-agent system that can be used for visualising simulated pedestrian activity and behaviour to support the assessment of design performance. This system is based on cellular automata and agent technology. Agents represent objects or people with their own behaviour, moving over a pedestrian network. Each agent is located in a simulated space, based on the cellular automata grid. Each iteration of the simulation is based on a parallel update of the agents conforming local rules. Agents positioned within an environment have sensors to perceive their local neighbourhood and affect their environment. In this manner, autonomous individuals and the interaction between them can be simulated by the system. 1 Introduction Architects and urban planners are often faced with the problem to assess how their design or planning decisions will affect the behaviour of individuals. Various performance indicators are related to the b...

We study several deterministic one-dimensional traffic models. For integer positions and velocities we find the typical high and low density phases separated by a simple transition. If positions and velocities are continuous variables the model shows self-organized criticality driven by the slowest car.