交通运输系统工程与信息 ›› 2025, Vol. 25 ›› Issue (5): 72-82.DOI: 10.16097/j.cnki.1009-6744.2025.05.006

• 自动驾驶与智慧交通 • 上一篇    下一篇

考虑心理因素的自动驾驶车用户停车选择行为分析

韩艳*1,袁昌银1,唐心恬1,关宏志1,2   

  1. 1. 北京工业大学,交通工程北京市重点实验室,北京100124;2. 新疆大学,新疆交通基础设施绿色建养与智慧交通管控重点实验室,乌鲁木齐830017
  • 收稿日期:2025-07-16 修回日期:2025-08-12 接受日期:2025-09-01 出版日期:2025-10-25 发布日期:2025-10-25
  • 作者简介:韩艳(1977—),女,江苏盐城人,副教授。
  • 基金资助:
    国家自然科学基金(71971005)。

Parking Choice Behavior Analysis of Autonomous Vehicle Users Considering Psychological Factors

HAN Yan*1, YUAN Changyin1, TANG Xintian1, GUAN Hongzhi1,2   

  1. 1. Beijing Key Laboratory of Traffic Engineering, Beijing University of Technology, Beijing 100124, China; 2. Xinjiang Key Laboratory of Green Construction and Smart Traffic Control of Transportation Infrastructure, Xinjiang University, Urumqi 830017, China
  • Received:2025-07-16 Revised:2025-08-12 Accepted:2025-09-01 Online:2025-10-25 Published:2025-10-25
  • Supported by:
    National Natural Science Foundation of China (71971005)。

摘要: 具备远程自动代客泊车功能的自动驾驶车将用户送到目的地后,可自主空驶至远端停车场停车,为解决热点区域停车泊位不足等问题提供了新的方法,同时也带来新的选择问题。本文基于消费者购买决策理论,引入感知风险和等待态度两个心理潜变量,挖掘自动驾驶用户停车选择机理,设计并实施了自动驾驶用户停车选择行为实证调查,构建集成心理潜变量的停车选择行为(ICLV)模型与不考虑心理潜变量的多项Logit模型,并进行敏感性分析。研究结果表明:心理潜变量、个人信息属性、出行属性和停车方案属性均对用户选择停车位置具有显著影响,与传统的多项Logit模型相比,ICLV模型拟合度显著提升。目的地停车费对停车方案选择概率变化富有弹性,当目的地停车费从11元·h-1增加至15元·h-1时,目的地停车的选择概率从59.9%下降至34.4%,近端停车、远端停车的选择概率分别从11.8%、28.3%上升至19.4%、46.2%。当用户的感知风险厌恶度和等待厌恶度分别由1上升至5时,用户选择目的地停车的概率分别由12.0%、24.4%上升至53.3%、52.0%。研究结论可为自动驾驶时代区域停车场差异化停车定价提供理论依据。

关键词: 城市交通, 停车选择行为, 混合选择模型, 自动驾驶汽车, 心理因素

Abstract: Autonomous vehicles equipped with long-range autonomous valet parking function drop off the users at their destinations and then autonomously idle to the remote parking lot to park. It can present an innovative solution to the spatial and temporal misallocation of parking resources in urban centers, and at the same time bring a new problem of choice. Based on the theory of consumer purchase decision, the parking choice mechanism of Privately-owned Automated Vehicles (PAVs) is discussed by introducing two psychological latent variables: perceived risk and waiting attitude. An empirical survey on the parking choice behavior of PAVs users was designed. And an Integrated Choice and Latent Variable (ICLV) model and MNL model were developed for the parking choice behavior of AV users. The results indicate that psychological latent variables, along with personal characteristics, travel attributes, and parking program attributes, significantly influence the parking decisions of users. Moreover, the ICLV model, which incorporates psychological latent variables, demonstrates a notably superior fit over the traditional multinomial Logit model. Destination parking fee is elastic to changes in the probability of parking choice. When the destination parking fee increases from 11 yuan·h-1 to 15 yuan·h-1, the probability of parking choice will decrease from 59.9% to 34.4%, and the probability of proximal parking slots and remote parking slots will increase from 11.8% and 28.3% to 19.4% and 46.2%, respectively. The probability of choosing destination parking will increase from 12.0% and 24.4% to 53.3% and 52.0% when the risk and pick-up waiting aversions perceived by users increase from 1 to 5, respectively. The research findings can provide a theoretical basis for differential parking pricing in regional parking lots in the era of autonomous vehicles.

Key words: urban traffic, parking choice behavior, hybrid choice model, autonomous vehicles, psychological factors

中图分类号: