A simple and flexible methodology is proposed in order to define Power-Two Wheelers (PTWs) riding profiles by distinguishing between regular and irregular PTW behaviors, by using high resolution naturalistic riding data. “Irregularities” in riding behavior are consistently expressed as outlying values in the multivariate consideration of a set of riding parameters. The detected irregularities are those that diverge from the centroid of the jointly considered riding variables and define critical riding situations that are further associated to typical riding events. Results indicate that the joint consideration of variables that are directly connected to the mechanical characteristics of the PTW, such as breaking, wheel speed, throttle and steering, are adequate to distinguish the regular from irregular riding behavior. Moreover, a regressor is constructed using neural networks and the influential determinants to the deviation from the rider’s regular actions are evaluated.
|Tags||driver behaviour, motorcyclists, statistical modelling, traffic management|