Knowledge-based approach to modeling urban dynamics

Abstract

The model representing the complexity of the pedestrian mobility has to incorporate the nature of the modeled phenomenon by accounting the interdependence between human behavior and urban environment. Our efforts are directed towards correlating emergent behavior patterns of different types of pedestrians to contextual knowledge that will help us map realistic pedestrian behavior into agent’s decision making capabilities. We propose that agent’s beliefs, goals and decision-making strategies should be derived directly from the integrated urban knowledge. Causal probabilistic models that are based on Bayesian inference are proposed as a potential solution to some of the challenges in the pedestrian agent modeling.