| In the preceding decades,autonomous vehicles(AVs)and emerging transportation network companies(TNCs)have gradually changed the travelers’ travel modes and parking choices.With the help of artificial intelligence algorithms and hardware equipment such as sensors and chips,autonomous driving technology has developed rapidly.Travelers can be relieved from the role of the private car drivers and be transformed into passengers in AVs,so they can enjoy leisure time or devote to work during travel.Furthermore,the traveler does not need to park the autonomous vehicle near the final destination.Instead,the AVs will drop off travelers at the final destination,and then AVs park themselves in the nearest parking spots or drive themselves to find other cheaper parking spaces.This thesis starts with evaluating the impact of facility layout for AVs to deal with the opportunities and challenges.Chapters 2 and 3 focus on analyzing the internal layout of a single parking lot for AVs.Chapter 4 considers TNCs pick-up/drop-off passengers and the space allocation optimization of curbside resources.Chapter 5 analyzes the location and allocation of parking facilities in urban areas and explains how autonomous driving and autonomous parking technologies will change parking from multiple dimensions.Moreover,try to answer the more profound question:how AVs will change the urban geographic space?The research main contents and innovations of this paper are summarized as follows:·Research on the optimization of the spatial layout of the parking.First of all,it can be found that the size of autonomous driving parking spaces in the existing research is assumed to be a fixed minimum size.However,the actual AV rarely reforms its mechanical structure,and the constraint of the mechanical steering in the parking process is more practical when calculating the required parking space.Therefore,in this paper,based on the Ackerman steering principle,a mixed-integer nonlinear programming model of parking capacity in the parking area was established for the first time.This model weighs the trade-off between parking space width and parking aisle width.Previous studies have assumed that autonomous vehicles can perfectly fit in the space,ignoring the impact of specific parking strategies.In this study,three types of parking strategies are considered:front-in,reverse-in,and translation,and analyze the difference in space utilization.By comparing the traditional manual parking facility,it is concluded that the vast majority of AVs’ parking space efficiency gains can be obtained under current automotive engineering practice in which only the front two wheels pivot.·For the problem of optimizing the allocation of the parking zone and pick-up/drop-off zone for AVs parking facility,we consider the possible impact of the time efficiency of the pick-up/drop-off zone for the whole parking system.Inspired by queueing theory,we define the process of AV passengers waiting for vehicles to pick up passengers in the pick-up/dropoff zone.The optimization goal is to establish a model to maximize the total revenue of the vehicle storage zone and the pick-up/drop-off zone and weigh the income from economic activities during the parking period and the cost of allocating space for picking up and dropping off for time efficiency in a parking facility.·Motivated by the rapid growth in new types of mobility-related activities seeking to use curbside space in cities,this paper proposes a framework for modeling competing demands to use the curbside.We extend the classical Bid-Rent Theory of urban land use into the curbside’s new context,the interface between the transportation network and adjacent urban land uses.The framework is intended to allow curbside space to be optimally allocated transparently,in contrast to the current practice of ad-hoc heuristic decision-making about curbside regulation.In the bi-level model,a curb space manager and individual travelers both make mutually interdependent choices.In a set of numerical analyses of a simple case study network,we demonstrate that the proposed framework is tractable,is flexible to simulating various types of curbside uses with different properties,and is capable of exhibiting intuitive sensitivity to systematic variation in inputs.· An optimization model for the location of remote parking facilities for AVs is proposed.Due to the difference in space occupation and parking methods from traditional parking,there are considerable differences in the choice preferences and cost factors of "remote parking".In the past,research methods lacked the consideration that the site selection layout adversely affected the commuter’s parking behavior,resulting in the inability to evaluate the optimal layout solution from a system perspective.Therefore,to comprehensively consider the impact of the remote parking strategy on the parking system and urban spatial geography.This chapter presents a model to study commuting parking facilities’ location,number and capacity for AVs.In the parking system model,city planners determine the location,number and capacity of remote parking facilities to reduce the total cost.The commuter parking choice considers travelers’ parking choice utility function.A simulation based on Monte Carlo method is designed to obtain the approximate optimal solution of the model.The framework aims to analyze the optimal allocation and layout of remote parking facilities to minimize the system’s total cost.This research helps to deepen the understanding of spatial layout and urban geography under the influence of transportation innovation.It provides the theoretical basis and methodological support for detailed parking planning,management,and improve parking services. |
