Delays, Inaccuracies and Anticipation in Microscopic Traffic Models

Preprint
Publisher Website
Google Scholar
  • 9 March 2005
Abstract
We generalize a wide class of time-continuous microscopic traffic models to include essential aspects of traffic dynamics not captured by these models, but relevant for human drivers. Specifically, we consider (i) finite reaction times and limited attention, (ii) estimation errors, (iii) looking several vehicles ahead (spatial anticipation), (iv) temporal anticipation, and (v) long-term adaptation to the global traffic situation. The estimation errors are modelled as stochastic Wiener processes and lead to time-correlated fluctuations of the acceleration. We have applied these generalizations to the intelligent-driver model and show that the destabilizing effects of reaction times, limited attention, and estimation errors can essentially be compensated for by spatial and temporal anticipation, that is, the combination of stabilizing and destabilizing effects result in the same qualitative macroscopic dynamics as that of the underlying simple car-following model which, in many cases, justifies the use of the latter. While the qualitative dynamics is unchanged, the effects of multi-anticipation increase both spatial and temporal scales of complex patterns of congested traffic such as stop-and-go waves and smear out the transition zones between free and congested traffic states, in agreement with real traffic data. Remarkably, the anticipation mechanisms allow accident-free smooth driving in complex traffic situations even if reaction times exceed typical time headways.

This publication has 0 references indexed in Scilit: