A Proposed Method for Investigating the Effects of Friction Supply on Driver Speed Selection on Horizontal Curves

Open Access
Topper, Stefan Andreas
Graduate Program:
Mechanical Engineering
Master of Science
Document Type:
Master Thesis
Date of Defense:
July 06, 2018
Committee Members:
  • Sean N Brennan, Thesis Advisor
  • Matthew B Parkinson, Thesis Advisor
  • Hosam Kadry Fathy, Committee Member
  • Karen Ann Thole, Committee Member
  • Friction
  • Vehicle Dynamics
  • Risk Perception
  • Horizontal Curves
  • Driving Behavior
This thesis proposes an experimental method which can be used to study how changes to the friction supply available between a vehicle’s tires and the road surface affect a person’s driving behavior on horizontal curves. The motivation behind this experiment is the theory that automobile drivers have internal risk models which determine an acceptable level of side friction while the vehicle is performing a cornering maneuver. While previous research has aimed to define this acceptable level of side friction in order to improve roadway design, the relationship between this value and the friction supply has not been studied extensively. In order to develop an appropriate experimental method, two pilot tests were conducted at The Pennsylvania State University Larson Transportation Institute Test Track. These pilot tests involved in-vehicle driving experiments during which test participants sat in the passenger seat of a mapping van equipped with a differential global positioning system (DGPS) and an inertial measurement unit (IMU) for data collection. Each participant was driven by a licensed driver for tests on a dry asphalt surface and a wet asphalt surface. The study was split into two test protocols: the latent response protocol and the learned response protocol. For the latent response protocol, the vehicle approached a series of traffic cones at a predetermined constant speed. Each cone represented a potential turn radius, and the participants were required to select an appropriate cone for the given driving speed. However, during the latent response protocol, the participants were never actually exposed to the turning maneuver. During the learned response protocol, one turn radius was presented to the participant at a time with a traffic cone, and he or she had to select an appropriate driving speed to make this turn. In this case, the driver would turn the steering wheel 270 degrees and maintain the participant’s selected speed in an attempt to make the desired turn. The results of the pilot tests indicate that people are much better at selecting an appropriate driving speed for a specific turn than they are at selecting a turn radius for a given driving speed. The participants’ responses also revealed several aspects of the experiment which could be improved. This led to the proposal of an altered experimental method which focuses more on naturalistic driving behavior and uses turn radius indicators more similar to driving lanes on a roadway. In combination with future studies involving the use of a driving simulator, this proposed experiment could help to identify the risk cues related to friction supply that drivers use to determine their driving behavior on horizontal curves.