Traffic Simulation in Traffic Intersection Using Improved Krauss and Improved Mobil Algorithm Implemented in Python

Maria April Rose Andaca and Jerwin Lloyd Cruda

Adviser: Adrian Roy Valdez, PhD

Abstract

The terrible situation of traffic in the Philippines demands better implementations of traffic rules and routing. As a result, new traffic regulations are being proposed and tested on road. Bad proposals tested on-road, however, results in a worse traffic condition that affects drivers and passengers alike.

A traffic computer simulator is created to minimize on-road testing and speed up gathering of information for various traffic proposals. Current traffic models for car following and lane changing were improved to take into account Filipino traffic behaviors. A new model, which uses time as basis of comparison, were also created to handle motions on traffic intersections. These were all implemented in Python 2.7. A crude test was done which showed that the models follow a no-collision assertion. Statistical testing and comparison on actual traffic data should be completed to affirm the validity of the said models in the real world.