Highlights This paper invents the “automated pedestrian shuttle” (APS) to transport pedestrians at an autonomous intersection. This paper models the optimum route choices for APSs based on virtual node technique. This paper captures the conflicts between APSs and AVs by discretizing the intersection. This paper explores the optimum entering time for each APS and AV to minimize the overall passenger delay.
Abstract This paper is among the first to address the problem of pedestrians’ crossing at an autonomous intersection without traffic lights. To guarantee pedestrians’ safety, all the pedestrians who intend to cross the street should firstly board into a fully enclosed automated transport unit, which is referred to as an “automated pedestrian shuttle” (APS) in this paper, then the APS carries the pedestrians across the street. Because APSs can shuttle back and forth through the intersection, the routing problem for APSs is modeled first. Then a Mixed Integer Linear Program (MILP) based AIM model is proposed to simultaneously optimize for the time of APSs and upcoming vehicles to enter the intersection. A numerical analysis is conducted to examine the performance and effectiveness of the proposed methods.
Autonomous intersection management with pedestrians crossing
2021-12-07
Article (Journal)
Electronic Resource
English
Centralized MPC for autonomous intersection crossing
IEEE | 2016
|Trans Tech Publications | 2014
|Child pedestrians' crossing gap thresholds
Online Contents | 1998
|