Emmanouil (Manos) Barmpounakis defended his PhD dissertation on the decision-making process of drivers during overtaking by Powered Two Wheelers, July 2017

Emmanouil (Manos) Barmpounakis has successfully defended his PhD dissertation titled: “Investigating the decision-making process of drivers during overtaking by Powered Two Wheelers”, on July 13th 2017. This PhD thesis was carried out at the Department of Transportation Planning and Engineering at the School of Civil Engineering of the National Technical University of Athens under the supervision of Prof. John Golias.

The aim of the specific PhD thesis is to investigate the decision making process of drivers during overtaking by Powered Two Wheelers and their interaction with the rest of the traffic in an urban road arterial.

In order to create a detailed naturalistic PTW trajectory database, the innovative use of Unmanned Aerial Systems (UAS) was preferred the specific database paved the way for advanced methodological tools. Decision Trees and other meta-algorithms, a special case of Machine Learning (ML) models, are developed in order to model the unconventional overtaking patterns of PTW drivers. The developed decision tree models enable the identification of the significant factors during overtaking such as speed difference, spatial factors and vehicle type and positions Then, the interactions between PTW and the rest of the traffic are examined using the concepts of Game Theory. Results show that cooperation can emerge, even in a competitive environment as the roads are.

The contribution of results and findings from the modeling of driver decision-making and the interaction of different types of vehicles is of particular importance for the design of modern Cooperative ITS, particularly in an environment where V2V technologies, autonomous and Connected Vehicles are among the hot topics of the international research community.

Summary 

Presentation: ppt