Shared Autonomous Vehicles Research Articles

A joint behavioral choice model for adoption of automated vehicle ride sourcing and carsharing technologies: Role of built environment & sustainable travel behaviors

Rapid advancements in vehicle automation and a shift towards collaborative consumption trends will disrupt future urban mobility systems. We present a joint model of consumers’ affinity towards shared automated vehicles (SAVs) with two distinct yet related configurations: automated vehicle (AV) carsharing and AV ride sourcing. Compact and walkable built environment is known to support active travel and shared mobility but the potential impacts of built environment on adoption of automated shared mobility are largely unknown. Besides sociodemographic, sustainable travel, and technology-related factors, we examine the role of neighborhood-level walkability (composite measure of land-use, pedestrian-oriented urban design, and transit accessibility) in shaping public’s opinion about AV carsharing and AV ride sourcing technologies. Detailed travel behavior and stated preference/taste data for over 2700 individuals from the most recent 2019 Puget Sound Travel Survey are complemented with neighborhood-level high-resolution data on built environment fabric. Two dependent variables of interest include individual’s level of interest in AV carsharing and AV ride sourcing programs. A novel and comprehensive geo-additive heterogeneous copula-based joint behavioral model is developed - revealing complex stochastic dependence patterns between users’ interest in AV ride sourcing and carsharing programs and different layers of systematic and random taste heterogeneity. Results show that improving neighborhood walkability can be an effective way to increase public’s interest in using automated ride sourcing and carsharing services. Findings also highlight the opportunity to enable SAV adoption by harnessing the synergies between green (transit) travel, technology use for work/non-work purposes, use of existing mobility-on-demand (MOD) services, and affinity towards automated MOD services. Results are translated to potential policies for designing proactive behavioral and built environment interventions for accelerating SAV adoption with an enhanced sensitivity towards the needs of distinct and more vulnerable population groups (e.g., older, unemployed, and/or less educated individuals). The study methodologically contributes by presenting a comprehensive behavioral model for SAV adoption by differentiating ride sourcing and carsharing.

Evaluating the Impacts of Optimization Horizon on the Shared Autonomous Vehicle Reservation Request System

With the development of technology, shared autonomous vehicles may become one of the main traffic modes in the future. Especially, shared autonomous vehicle reservation system, commuting, and other trips with fixed departure time mostly submit their travel requests in advance. Therefore, it is important to reasonably match shared autonomous vehicles and reservation demands. In this paper, reservation requests are divided into short-term and long-term requests by inputting requests in a more realistic way. An integer linear programming model considering operator scheduling cost and system service level is established. A detailed scheme considering rolling horizon continuity and ridesharing is used to improve the dispatching result. Based on traffic data in Delft, the Netherlands, 164 scenarios are tested in which the parking cost, fuel cost, ridesharing effect, service level, and network size are analyzed. The results show that a better relocation and ridesharing matching scheme can be obtained when the rolling horizon is small, while the overall effect is better when the rolling horizon is large. Moreover, the buffer time, distance, and travel time limit for vehicle relocation should be selected according to the request quantity and the calculation time requirement. The result can provide a suggestion for the dispatching of shared autonomous vehicle reservation system with ridesharing.

Modal Shift and Shared Automated Demand-Responsive Transport: A Case Study of Jerusalem

By definition, the Mobility-as-a-Service (MaaS) system integrates transportation modes of very different flexibility - taxis, buses, light rail, ride-hailing. MaaS major unknown is the effectiveness of the ride-sharing modes. We employ a calibrated and validated MATSim multi-modal traffic model of the Jerusalem Metropolitan Area (JMA) to assess the introduction of Shared Autonomous Vehicles (SAV), as a possible game-changer of the existing equilibrium between the Public Transport (PT) and private cars. First, we confirm the recent empirical observations that ride-sharing modes mostly attract PT users, while their attractiveness for private car users is relatively low. Second, we investigate the problem of preserving travelers’ flows to the Jerusalem center while reducing personal car use. For this purpose, we investigate the effect of parking prices and congestion charges in the system with the additional SAV fleet that serves trips to-and-from the center of the city. We propose a balanced set of carrot-and-stick measures to enforce a sustainable modal shift in JMA towards the use of PT and SAV service.

The Relationship between Exposure to and Trust in Automated Transport Technologies and Intention to Use a Shared Autonomous Vehicle

The Relationship between Exposure to and Trust in Automated Transport Technologies and Intention to Use a Shared Autonomous Vehicle

BigGAN evaluation for the generation of vehicle interior images

The number of shared autonomous vehicles (SAV) tends to increase in the coming years, highlighting the need to create monitoring systems that safeguard the integrity of the SAV and the safety of passengers. For the creation of monitoring systems, it is necessary to develop algorithms capable of detecting and classifying a multitude of objects (i.e. dangerous, forgotten, damaged), in different types of vehicles. Currently, deep learning (DL) algorithms present themselves as the best option to solve this problem, but require a large amount of data for training. This article focuses on the use of Generative Adversarial Networks (GAN) for the automatic generation of artificial images of vehicle interiors. Specifically, we propose to employ the BigGAN arquitecture, with the combined implementation of two recent techniques that aim to improve training stability and GAN generalization, namely consistency regularization and differential augmentation. With an expanded version of MoLa-VI dataset (made publicly available), satisfactory results were obtained with the proposed. Moreover, CR+BigGAN combination presented the best results, achieving a Fréchet Inception Distance of 28.23 and an Inception Score of 17.19.

Optimized in-vehicle multi person human body pose detection

The number of Shared Autonomous Vehicles (SAV) will increase in the coming years. The absence of human driver will create a new paradigm for in-car safety. This paper addresses this problem by presenting an approach to estimate the human body pose inside a vehicle. We propose to use a customized version of the OpenPose framework, to perform the task of human body pose detection for the front passengers inside a vehicle. The OpenPose method was been evaluated with three different backbones: VGG19, MobileNetV1 and MobileNetV2, using different hyperparameters and ablation scenarios. Moreover, synthetic images were used, which simulate a depth sensor perspective from the center of the front seats. The dataset is comprised by images with 1 and 2 passengers, from 18 different subjects inside of 7 different vehicles, thus making a total of 45360 different images. The OpenPose method with the MobileNetV2 backbone showed the most efficient results between precision and performance, achieving a mean Average Precision (mAP) of 90%, Area Under ROC Curve (AUC) of 73%, and 0.0189 seconds per image (s/img).

Cost Analysis of Shared Autonomous Electric Vehicles with Charging Infrastructure

Cost Analysis of Shared Autonomous Electric Vehicles with Charging Infrastructure

Research on Parking Choice Behavior of Shared Autonomous Vehicle Services by Measuring Users’ Intention of Usage

Research on Parking Choice Behavior of Shared Autonomous Vehicle Services by Measuring Users’ Intention of Usage

Required Simulated Population Ratios for Valid Assessment of Shared Autonomous Vehicles’ Impact Using Agent-Based Models

Required Simulated Population Ratios for Valid Assessment of Shared Autonomous Vehicles’ Impact Using Agent-Based Models

Social Adoption of the Shared Autonomous Vehicle and Multimodal Equilibrium in a Continuous and Multiplayer Game-Based Experiment

Social Adoption of the Shared Autonomous Vehicle and Multimodal Equilibrium in a Continuous and Multiplayer Game-Based Experiment

An Integrated Framework for Locating Depots in Shared Autonomous Vehicle Systems

An Integrated Framework for Locating Depots in Shared Autonomous Vehicle Systems

An Integrated Personalized Incentive Scheme for Shared Autonomous Vehicles

An Integrated Personalized Incentive Scheme for Shared Autonomous Vehicles

A Systemic View of Future Mobility Scenario Impacts on and Their Implications for City Organizational LCA: The Case of Autonomous Driving in Vienna

Autonomous vehicles (AV) are expected to significantly reshape urban mobility. Whether advancements at vehicle level also translate into positive environmental outcomes at city level is still uncertain. We investigate under which conditions a city could enable low emission AV mobility and what challenges are to be expected along the way from an environmental point of view. We build upon our recent environmental performance study of Vienna and combine city organizational life cycle assessment (city-OLCA) with AV transport models from literature for three AV use cases: an own AV, a shared AV, and a shared AV ride service. Most cases lower Vienna’s passenger capacity (by up to 28%) and increase motorized road traffic by a maximum of 49% (own AVs). Traffic relief is observed for shared AVs (−40%) if accompanied by a conventional car ban. This case reduces transport related GHG emissions compared to both Vienna’s current baseline (−60%) and a future electrified transportation system (−4.2%). These transformations have also shifted emission responsibility to the public level. While Vienna’s total GHG emissions could be reduced by 12%, the city’s emission responsibility increases from 25% to 32%. Efficient mass transit, the electrification of the mobility sector and grid decarbonization are key to reducing transport emissions in Vienna. The direction of GHG emission development will be determined by the extent to which these conditions are promoted. AV mobility probably will not be a main contributor.

Sustainable Renewal of Underground Parking Space in the Scenario of Shared Autonomous Vehicles

Shared autonomous vehicles (SAVs) are a major development direction in international scientific and technological innovation. One of the most popular features of SAVs in the urban space is that they can significantly reduce the need for parking. The urban underground parking space (UPS) is currently the largest static traffic space, especially in high-density urban centers. Under the SAV scenario, the need for the renewal of UPS will increase in the near future. However, renewal of the UPS is difficult due to its special form features, which are greatly restricted by the external environment, thus necessitating targeted methods and strategies. This research first conducted field investigations and data collection on the spatial morphology and service conditions of typical UPSs in different areas of Hangzhou city. Based on the driver status response and the multi-objective attribute models, the time-series evaluation method and function replacement decision model for the sustainable renewal of underground parking were established. The research also discusses appropriate design strategies for the combination of spatial characteristics and functional replacement goals of typical samples. The conclusions will provide scientific guidance for the future design practice of architects and urban designers in SAV.

Predicting Autonomous Shuttle Acceptance in Older Drivers Based on Technology Readiness/Use/Barriers, Life Space, Driving Habits, and Cognition

Shared autonomous vehicle services (i. e., automated shuttles, AS) are being deployed globally and may improve older adults (>65 years old) mobility, independence, and participation in the community. However, AS must be user friendly and provide safety benefits if older drivers are to accept and adopt this technology. Current potential barriers to their acceptance of AS include a lack of trust in the systems and hesitation to adopt emerging technology. Technology readiness, perceived ease of use, perceived barriers, and intention to use the technology, are particularly important constructs to consider in older adults' acceptance and adoption practices of AS. Likewise, person factors, i.e., age, life space mobility, driving habits, and cognition predict driving safety among older drivers. However, we are not sure if and how these factors may also predict older adults' intention to use the AS. In the current study, we examined responses from 104 older drivers (Mage = 74.3, SDage = 5.9) who completed the Automated Vehicle User Perception Survey (AVUPS) before and after riding in an on-road automated shuttle (EasyMile EZ10). The study participants also provided information through the Technology Readiness Index, Technology Acceptance Measure, Life Space Questionnaire, Driving Habits Questionnaire, Trail-making Test Part A and Part B (TMT A and TMT B). Older drivers' age, cognitive scores (i.e., TMT B), driving habits (i.e., crashes and/or citations, exposure, and difficulty of driving) and life space (i.e., how far older adults venture from their primary dwelling) were entered into four models to predict their acceptance of AVs—operationalized according to the subscales (i.e., intention to use, perceived barriers, and well-being) and the total acceptance score of the AVUPS. Next, a partial least squares structural equation model (PLS-SEM) elucidated the relationships between, technology readiness, perceived ease of use, barriers to AV acceptance, life space, crashes and/or citations, driving exposure, driving difficulty, cognition, and intention to use AS. The regression models indicated that neither age nor cognition (TMT B) significantly predicted older drivers' perceptions of AVs; but their self-reported driving difficulty (p = 0.019) predicted their intention to use AVs: R2 = 6.18%, F (2,101) = 4.554, p = 0.040. Therefore, intention to use was the dependent variable in the subsequent PLS-SEM. Findings from the PLS-SEM (R2 = 0.467) indicated the only statistically significant predictors of intention to use were technology readiness (β = 0.247, CI = 0.087-0.411) and barriers to AV acceptance (β = −0.504, CI = 0.285-0.692). These novel findings provide evidence suggesting that technology readiness and barriers must be better understood if older drivers are to accept and adopt AS.

Multitasking onboard of conventional transport modes and shared autonomous vehicles

To mitigate the negative impact of travel time, travelers are predisposed to multitask onboard when traveling. The preferences of people determine how the travel time is utilized. Travelers try to minimize the unused part of travel time or convert it into a productive time by doing onboard activities and choosing the proper transport mode. Current study focuses on the travelers’ behavior onboard when traveling to their main destinations. The research studies the urban areas where the travel time is relatively short. Generally, most studies examine multitasking onboard of one or more conventional transport modes (CTMs) without defining the area type of the trips. Moreover, there is a demand on those papers that concern multitasking onboard of autonomous vehicles (AVs). Thus, current research focuses on one case of AVs while studying onboard multitasking, i.e., the shared autonomous vehicle (SAV). A survey is designed and distributed in Budapest, Hungary, and a sample size of 276 travelers is collected. Respondents choose the onboard activities and the tools that are carried and used onboard during their travel. The onboard activities are examined with SAV. During the analysis, the following methods are used: multiple response analysis, central tendency, Chi-square, Cramer's V, Cochran's Q test, and A Kruskal-Wallis. As a result of analyzing the differences between the activities onboard of SAVs and CTMs, the factors that affect onboard activities are found. Factors influencing the usage of the travelers’ carried tools and those that negatively affect the travel time and the acceptability of SAVs are identified, too. In conclusion, travelers’ onboard behavior is studied, and it is found that SAVs are preferred over CTMs to maximize the onboard activities and consequentially, increase the utilities.

One-to-many matching and section-based formulation of autonomous ridesharing equilibrium

This paper models autonomous ridesharing — multiple travelers simultaneously riding one shared autonomous vehicle (SAV) — in a network equilibrium setting with mixed SAV and human-driven vehicle (HV) traffic. We make two major methodological contributions. First, a novel one (SAV)-to-many (riders) matching is proposed to characterize the waiting times of an SAV and multiple travelers who share rides in the SAV during online matching, which is a nontrivial generalization of the one-to-one matching without ridesharing. Our matching characterization considers the possibilities of a traveler matched with an SAV starting from the same origin, whereto the SAV moved unoccupied as a result of either pickup or relocation, or with an in-service SAV that goes through the traveler's origin. Second, a section-based formulation for SAV ridesharing user equilibrium is introduced to characterize the SAV traveler flow, which accommodates the possibility that an SAV traveler's itinerary (OD pair) is different from that of the serving SAV and other travelers in the SAV. Unlike the existing link and route based ridesharing formulations, the notion of section both prevents undesired traveler en-route transfer(s) and allows travelers of multiple ODs to share rides, meanwhile respecting the SAV seat capacity constraint. In addition to the above two methodological contributions, the optimal SAV fleet size, fare, routing, and allocation (to in-service, pickup, and relocation states) decisions of a transportation network company (TNC) are formulated. The TNC decisions anticipate traveler reactions as characterized by a new multimodal autonomous ridesharing user equilibrium (MARUE), which is put forward with a proof of its existence and finds the endogenous market shares and road congestion effects of SAV/HV. Original insights are obtained from model implementation, including substantial systemwide benefit of ridesharing, marginal benefit of relocation in the presence of ridesharing, and diminishing economies of SAV size.

Shared automated vehicle fleet operations for first-mile last-mile transit connections with dynamic pooling

Shared automated vehicles (SAVs) have the potential to promote transit ridership by providing efficient first-mile last-mile (FMLM) connections through reduced operational costs to fleet providers and lower out-of-pocket costs to riders. To help plan for a future of integrated mobility, this paper investigates the impacts of SAVs serving FMLM connections, as a mode that provides flexibility in access/egress decisions and is well coordinated with train station schedules. To achieve this objective, a novel dynamic pooling algorithm was introduced to match SAVs with riders while coordinating the riders' arrival times at the light-rail station to a known train schedule. Microsimulations of SAVs and travelers throughout two central Austin neighborhoods show how larger service areas, higher levels of SAV demand, and longer arrival times between successive trains require larger SAV fleet sizes and higher SAV utilization rates to deliver close traveler wait times. Four-person SAVs appear to perform similar to 6-seat SAVs but will cost less to provide. Using a dynamic pooling algorithm tightly coordinated with train arrivals (every 15 min) delivers 87% of travelers to their stations in time to catch the next train, while uncoordinated assignments deliver just 58% of travelers in time.

Modeling and Optimization of Multiaction Dynamic Dispatching Problem for Shared Autonomous Electric Vehicles

The fusion of electricity, automation, and sharing is forming a new Autonomous Mobility-on-Demand (AMoD) system in current urban transportation, in which the Shared Autonomous Electric Vehicles (SAEVs) are a fleet to execute delivery, parking, recharging, and repositioning tasks automatically. To model the decision-making process of AMoD system and optimize multiaction dynamic dispatching of SAEVs over a long horizon, the dispatching problem of SAEVs is modeled according to Markov Decision Process (MDP) at first. Then two optimization models from short-sighted view and farsighted view based on combinatorial optimization theory are built, respectively. The former focuses on the instant and single-step reward, while the latter aims at the accumulative and multistep return. After that, the Kuhn–Munkres algorithm is set as the baseline method to solve the first model to achieve optimal multiaction allocation instructions for SAEVs, and the combination of deep Q-learning algorithm and Kuhn–Munkres algorithm is designed to solve the second model to realize the global optimization. Finally, a toy example, a macrosimulation of 1 month, and a microsimulation of 6 hours based on actual historical operation data are conducted. Results show that (1) the Kuhn–Munkres algorithm ensures the computational effectiveness in the large-scale real-time application of the AMoD system; (2) the second optimization model considering long-term return can decrease average user waiting time and achieve a 2.78% increase in total revenue compared with the first model; (3) and integrating combinatorial optimization theory with reinforcement learning theory is a perfect package for solving the multiaction dynamic dispatching problem of SAEVs.

Travel Time Impacts of Using Shared Automated Vehicles along a Fixed-Route Transit Corridor

Shared automated vehicles (SAVs) offering a fixed-route transit may compete well against privately operated vehicles. This paper analyzes the system costs of all travelers along a 6.4-kilometer (4-mile) corridor under different penetration rates for 10-seat SAVs. The work prices out walking, waiting, riding, and driving times for all travelers in the corridor, along with vehicle ownership, parking, and operating costs. Results show that such self-driving mini-buses or SAVs lower total costs per passenger-kilometer traveled when SAV mode split exceeds 20 percent, even though walking and waiting are valued at relatively high cost. Such vehicles dramatically free up pavement (and parking) space, and perform even better when parking costs at drivers’ destinations are high.