Search Results

The improvement of road transport safety requires the development of advanced vehicle safety systems, whose development could be facilitated by using complex interaction models of different road users. To this end, this paper deals with the modeling of multi-vehicle/multi-pedestrian interactions at unsignalized crosswalks. This multi-agent modeling approach extends on the existing basic model covering only single-vehicle/single-pedestrian interactions. The basic model structure and parameters have remained the same, as it was previously experimentally calibrated and thoroughly verified. The proposed modeling procedure employs the basic model within the multi-agent setting based on its application to relevant single-vehicle and single-pedestrian pairs. The resulting, so-called pre-decisions are then used for making final crossing decisions in a current time step for each agent.

To enable smooth and low-risk autonomous driving in the presence of other road users, such as cyclists and pedestrians, appropriate predictive safe speed control strategies relying on accurate and robust prediction models should be employed. However, difficulties related to driving scene understanding and a wide variety of features influencing decisions of other road users significantly complexifies prediction tasks and related controls. This paper proposes a hierarchical neural network (NN)-based prediction model of pedestrian crossing behavior, which is aimed to be applied within an autonomous vehicle (AV) safe speed control strategy. Additionally, different single-level prediction models are presented and analyzed as well, to serve as baseline approaches.

A detailed experimental validation has been carried out to point to limitations of static wet-clutch friction model for typical clutch engagement transients. The model accuracy can be increased by incorporating the fluid film dynamics, as done in the lumped-parameter dynamic clutch model developed at the University of Purdue. That model is extended herein in order to increase its accuracy especially in the case of grooved clutches. The extensions include a description of clutch actuator dynamics and introduction of an empirical scaling factor for the fluid film thickness state equation. More rigorous treatment of fluid dynamics for the grooved clutch is also presented.