Data-Driven Approach for Modeling the Mixed Traffic Conditions Using Supervised Machine Learning
Narayana Raju (TU Delft - Transport and Planning)
Shriniwas Arkatkar ( Sardar Vallabhbhai National Institute of Technology)
Gaurang Joshi ( Sardar Vallabhbhai National Institute of Technology)
Constantinos Antoniou (Technische Universität München)
More Info
expand_more
Other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license such as Creative Commons.
Abstract
The article describes modeling vehicular movements using supervised machine learning algorithms with trajectory data from heterogeneous non-lane-based traffic conditions. The trajectory data on the mid-block road section of around 540 m is used in the study. Supervised machine learning algorithms are employed to model the vehicular positions. A set of parameters were identified for modeling the longitudinal and lateral positions. With the set of parameters, the algorithm’s potentiality for mimicking vehicular positions is evaluated. It was identified that supervised machine learning algorithms would model the vehicles’ positions with accuracy in the range of 20–60 mean absolute percentage error. The k-NN algorithm was marginally edging past all algorithms and acted as a promising candidate for modeling vehicular positions.