On Realizing the Shortest Time Strategy in a Ca Ff Pedestrian Dynamics Model

Abstract

A mathematical model of pedestrian movement on the basis of the stochastic cellular automata approach is proposed. A floor field (FF) model is taken as a basis model. The FF models imply that virtual people follow the shortest path strategy; meanwhile, they also follow the shortest time strategy. The focus of this study is mathematical formalization and model implementation of these features of pedestrian movement. Some results of the simulation are reported.

Calibrating floor field cellular automaton models for pedestrian dynamics by using likelihood function optimization

“…34 Ruggiero Lovreglio thanks also Dr. Giulio Santori (The University of Edinburgh) for his advice concerning optimization 35 problems. 36 …”

“…Hoogendoorn and Daamen [26] 35 introduced microscopic approach to estimate the NOMAD model by using MLE. Guo et al propose an approach based on the 36 LSE to calibrate their logit-type pedestrian model using sample data on the deviation angles, step velocities, and walking 37 speed-density relations obtained from experiments [13]. Another approach is that used by Guo et al [15] based on the 38 comparison between simulated evacuation times and observed evacuation times by using LSE.…”

“…2 shows 35 the result of these formulations. 36 This example does not include the influence of fixed obstacles. However, a pedestrian could prefer to not get too close 37 to an obstacle (e.g., a wall) given the so-called ''effective width'' principle [37].…”

“…For instance, different path strategies (i.e. shortest time versus shortest way strategy [36]) can be included 28 in Eq. (1).…”

Calibrating dynamic pedestrian route choice with an Extended Range Telepresence System

“…Travel time as determinant of the motion has been discussed and modeled in a number of previous works, sometimes for macro-or mesoscopic approaches [13,14,32,8,44], but more often within the scope of microscopic models. Of these the majority are field-based as the dynamic potential approach [12,49,46,26,27,23,47,29,28,7,24], but there exist also other methods which are based on an additional routing network, discrete choice or heuristics [5,3,11,21,34]. …”