An Open-loop Model of Bus Drivers' Lane Change Behavior Based on Naturalistic Driving Data

Accurately describing bus driver's behavior is crucial for driving safety intervention strategy. Based on naturalistic data collected from 6 bus drivers, a novel open-loop bus driver model is proposed, which describes bus driver's behavior in starting and ordinary lane change scenarios. 23...

Full description

Saved in:
Bibliographic Details
Main Authors: Huang, Xuyin, Pi, Dawei, Xie, Boyuan, Wang, Hongliang
Format: Conference Proceeding
Language:English
Subjects:
Analysis of variance
Analytical models
Bus Driver Model
Data mining
Data models
Driving Behavior
Lane Change
Parameter estimation
Parameter Fitting
Safety
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Accurately describing bus driver's behavior is crucial for driving safety intervention strategy. Based on naturalistic data collected from 6 bus drivers, a novel open-loop bus driver model is proposed, which describes bus driver's behavior in starting and ordinary lane change scenarios. 236 lane change segments classified into six groups are extracted, with analysis of variance method being applied to capture maneuver characteristics of bus drivers in typical lane change scenes. 50% cumulative frequency parameter and cluster centers generated by k-means method are used to learn parameters of the proposed model. The results show that the proposed driver model can well simulate behavior of bus drivers under natural starting and ordinary lane change scenes.
ISSN:1948-9447
DOI:10.1109/CCDC52312.2021.9601684