• Session No.77 Dynamics, Control and Safety of Two-wheeled Vehicles II -Motorcycles, Bicycles, and PMV- (OS)
  • May 23Room G40412:40-13:55
  • Chair: Tetsunori Haraguchi (Nagoya University)
Contents
In this session, we will collect research findings on the various dynamic characteristics of two-wheeled vehicles, rider characteristics, control, and safety, and discuss future directions.
Committee
Two-wheeled Vehicle Dynamics Committee
Organizer
Tetsunori Haraguchi (Nagoya University), Masaru Asakawa (Hitachi Astemo), Tomoya Kitani (Shizuoka University), Junji Hirasawa (National Institute of Technology (KOSEN), Ibaraki College)
For presentations that will not be available video streaming after congress, a “✕” is displayed in the “Video” column, so please check.
No. Video Title・Author (Affiliation)
1

Examination of the Appropriate Alignment of Omnidirectional Cameras in the Position Measurement System for a Motorcycle Using Image Processing

Junji Hirasawa (Ibaraki KOSEN(College of Tech.))

In this paper, a reexamination for the position measurement system is described, focusing on the number of cameras, the orientation of the optical axes, and the relative positions of the cameras. The position accuracy of the proposed method is assessed through verification experiments. Driving experiments are conducted to measure the locus for a motorcycle.

2

Analysis of the synchronization of the rider's posture, gaze, and maneuvers with the vehicle's motion.

Masakazu Tomosada・Masaru Katayama・Yoshihiro Fujioka・Yukito Fukushima・Daiki Izumi・Takeshi Kobuki・Syusuke Yamane・Akiyasu Takami (National Institute of Technology, Matsue College)

We have developed a system that can synchronously measure and analyze the rider's posture, gaze, and operation while riding and the vehicle's motion.
The results of riding analysis using this system will be reported.

3

A Proposal for Translating Vehicle Motion Sensing Data into Symbolic Sequences with Temporal Dynamics
-Hear the Motorcycle Sing-

Tomoya Kitani・Takuro Sone・Yotaro Yada (Shizuoka University)

This paper proposes a method for representing time-series motion sensing data as symbolic sequences that consider time scales .
It enables the computational analysis of actual physical sensing data from vehicle motion using machine learning and natural language processing techniques such as LLMs .
The results of this analysis can contribute to objectively assessing the experience of riders and the vehicle dynamics.

Back to Top