Abstract

Riding behavior is characterized by speed choices and evasive actions performed by riders while riding under normal conditions through naturalistic riding studies (NRS). Riders accelerate to attain free speeds and respond to traffic conflicts through various evasive actions. As safety of a motorized two-wheeler (MTW) rider predominantly depends upon his/her acceleration and deceleration behavior, we aim to classify acceleration and deceleration events obtained through a short-term NRS, and identify participants’ riding behaviors. The NRS included 58 riders and captured their kinematic profiles and surrounding traffic conditions through an instrumented MTW. Riders consisted of students and working-professionals with varying ages. Riders drove a 32 km round trip, and acceleration and deceleration events were extracted from riding profiles. The k-means clustering technique revealed three acceleration and three deceleration clusters. Acceleration clusters A1, A2, A3 were high-speed moderate-acceleration, moderate-speed low-acceleration and low speed high-acceleration clusters, respectively. Deceleration clusters D1, D2, D3 were low-speed high-deceleration, high-speed moderate-deceleration and moderate-speed low-deceleration clusters, respectively. This study revealed the most frequent riding behaviors (acceleration cluster A2 and deceleration cluster D3) which form MTW riders’ base-line riding behaviors. Principal component analysis revealed unique riding behavior of each acceleration and deceleration cluster. Study results also revealed significant differences between cluster contributions of student-riders and working-professionals, indicating that age and experience of riders also affect the number of acceleration/deceleration events.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call