01622558

Component

In the challenging urban environment, Powered Two Wheelers (PTW) are becoming more and more popular as a way of everyday commuting. Their “chaotic” manner to travel through traffic is systematically emphasized in the recent relevant literature, especially in relation to some observed complex maneuvers, such as overtaking. The aim of this paper is to model the overtaking patterns of PTW drivers using Decision Trees and other meta- algorithms to achieve enhanced performance. Based on detailed naturalistic trajectory data collected using Unmanned Aerial Vehicles in a three-lane arterial in Athens, Greece, two different models are developed. The first models the decision of the PTW driver to overtake or not the preceding vehicle, while the second models PTW driver’s intention to overtake or undertake (pass from the right) the preceding vehicle. The developed decision tree models enable the identification of the significant factors during overtaking. Finally, the applicability of the developed algorithms, as well as the importance of acquiring quality data using advanced equipment combined with advanced Machine Learning approaches are discussed.

This paper was sponsored by TRB committee ABJ70 Standing Committee on Artificial Intelligence and Advanced Computing Applications.

01618707

17-01614

English

16p

2017

Transportation Research Board 96th Annual Meeting

Digital/other

Figures; References; Tables

Algorithms; Decision trees; Motorcycles; Passing; Scooters; Two wheeled vehicles; Vehicle trajectories

Athens (Greece)

Data and Information Technology; Pedestrians and Bicyclists

Transportation Research Board Annual Meeting 2017 Paper #17-01614

TRIS, TRB, ATRI

Dec 8 2016 10:32AM