Automated Vehicles and People Living with a Disability: Opportunities, Challenges, and Future Directions for Sustainable Mobility

Abstract

This article reviews the current scientific literature that relates to automated vehicles (AVs), vehicles controlled by a computer instead of a human driver, and people living with a disability (PLWD). The aim of this review is (1) to explore how AVs might improve mobility for PLWD, (2) to identify research gaps to guide future studies, and (3) to examine the real-world applicability of existing research. A structured search following PRISMA guidelines identified 66 relevant peer-reviewed publications. The findings suggest that AVs hold promise in reducing transport-related social exclusion by increasing autonomy, flexibility, and accessibility for PLWD, thereby supporting the transition toward more inclusive and environmentally sustainable transport systems. However, the potential benefits of AVs for the mobility of PLWD depend on the type of vehicle considered (e.g., private vs. public transport) as well as the potential challenges related to the legal framework, accessibility standards, and addressing PLWD concerns, opinions, and needs. To overcome the existing obstacles to the widespread adoption of AVs and make them a real opportunity for PLWD, collaboration between all stakeholders in the sector (i.e., governments, industries, and disability associations) is needed. This review supports cross-sector collaboration for inclusive AV implementation.

1. Introduction

Disability affects many people around the world. The World Health Organisation estimates that around 16% of the population lives with a disability, representing 1.3 billion people [1]. Disability is recognized as resulting from the interaction between people with impairments and the environmental and attitudinal barriers that hinder their full and effective participation in society on an equal basis with others [2]. Impairments may affect physical, mental, intellectual, and/or sensory functions [2]. According to the Disability Creation Process Model [3], disability is situational and “inseparable from the consideration of environmental factors, society and community” (p. 27). In this model, a “disabling situation”, as opposed to “social participation”, relates to the hindering of the completion of life habits: daily activities and social roles. This hindering is influenced by the interaction between personal and environmental factors acting either as facilitators or as obstacles. Personal factors encompass identity factors, organic systems, and abilities, while environmental factors operate at different levels: micro (i.e., individual), meso (i.e., community), and macro (i.e., societal). As defined by the United Nations and the Disability Creation Process Model, disability is not the result of a medical condition, but rather an obstacle to social participation caused by the mismatch between the environment and an impairment.

In this context, restricted mobility is a major barrier to the social inclusion of people living with disabilities (PLWD) [4]. Mobility, defined here as both the means of transport available and the ability to access them [5], plays an essential role in accessing daily activities, such as professional activity, leisure, medical care, and social life. Accessible and inclusive transport fosters social inclusion, reduces inequalities, and is integral to creating sustainable, resilient communities. Sustainable development specifically calls for safe, affordable, and accessible transport for all, including PLWD [6].

Access to mobility can either alleviate or exacerbate existing social inequality, leading to “transport-related social exclusion” [7]. Social exclusion manifests as reduced access to and participation in different dimensions of social life, including health, economic, political, social, and cultural life [5]. In many countries, PLWD face significant barriers in using both public and private transportation, despite regulatory measures aiming to improve accessibility (e.g., [4,8,9,10,11,12,13]).

Currently, the transport infrastructure is still not completely accessible, and some PLWD cannot rely on public transport to move around (e.g., inadequate infrastructure or vehicles, unserved residential areas), meaning that some PLWD are still excluded from the current transport system. For example, 14.4% of PLWD in the USA reported lacking access to reliable transportation for daily living in 2022 (National Center for Health Statistics). As a result, PLWD might assess their social life as being restricted because of a lack of access to mobility [4,14] and travel less than the rest of the population [15]. These mobility-related issues are summarized in

Table 1. Summary of main mobility-related problems faced by people living with disabilities.

Category	Main Challenges	References
Infrastructure and Vehicles	Inaccessible transport systems and vehicles (e.g., lack of ramps, lifts, securement systems)	[4,8,9,10,11,12]
Geographic Coverage	Absence of public transport in certain residential or rural areas	[12,13]
Service Limitations	Inconsistent availability and performance of adapted or paratransit services	[8,14]
Trip Planning and Usability	Barriers in accessing and using booking systems, real-time info, or complex navigation routes	[5,7,9,10]
Dependence on Assistance	Strong reliance on caregivers or family for daily transport needs	[14,15]
Social Participation	Restricted participation in work, healthcare, and social activities due to poor mobility access	[5,7,13,14,15]

below.

Given these challenges, automated vehicles (AVs) are considered a potential solution to improve the mobility of PLWD (e.g., [16]) and contribute to achieving sustainable development, which promotes the development of safe, affordable, accessible, and sustainable transport systems for all [6]. AVs are classified by levels of automation ranging from 0 (fully manual) to 5 (fully autonomous) [17], starting with a conventional manual vehicle entirely operated by a human driver (Level 0). Higher levels of automation (Level 3 and above) allow complete delegation of vehicle control under specific conditions, with Level 5 AVs capable of fully autonomous travel. This technology has the potential to reduce dependency and increase independence for disabled people, but its effectiveness depends on the resolution of critical issues relating to accessibility, usability, and regulation.

The development of AVs is still an ongoing process, with a wide variety of experiments and demonstrations of vehicles equipped with Level 3 to 5 automated features (e.g., [18,19]). AVs might take the form of a personal vehicle, a shared vehicle operated in a carpooling fashion, or public transport vehicles operated on fixed or on-demand routes. In the present literature review, vehicles with levels of automation above Level 2 will be taken into consideration. Current research about automated driving systems above Level 2 mainly aims at designing AVs that are adapted to users’ needs, technically reliable, and economically viable (e.g., [20,21,22]). To develop AVs that are accessible to people with disabilities, it is essential to study their specific needs in relation to this technology from multiple angles (i.e., technical, economic, ergonomic, psychological, legal, governance, vehicle, and service design) to provide appropriate solutions. Although some research efforts, such as Dicianno et al. (2021) [23], address these issues from a broader perspective, considering both the elderly population and people with reduced mobility, the subject remains underexplored.

In this context, this paper aims to critically review the existing literature on AVs and PLWD to highlight the prerequisites necessary to ensure that AVs become a true mobility-enhancing innovation for this population. Specifically, this review seeks (1) to review the current literature with a focus on AVs and PLWD, highlighting the key prerequisites for this innovation to enhance mobility for this population, (2) to identify research gaps to guide future studies, and (3) to examine the real-world applicability of existing research. The methodology used to perform this literature review is first described; this is followed by the presentation of the literature review and the discussion of the results to identify key challenges and opportunities for future research.

2. Materials and Methods

Relevant literature was systematically selected from multiple academic databases to ensure a comprehensive and rigorous review. The selected databases included Google Scholar, Science Direct, PsycArticle, and PubMed. These databases were chosen for their broad coverage of interdisciplinary research on transportation, health, psychology, and engineering. The search was conducted in February 2025.

A structured search strategy was employed using a combination of carefully selected keywords to maximize the retrieval of relevant articles. The search terms included the following: (“automated vehicle” OR “self-driving car” OR “autonomous vehicle” OR “driverless car”) AND (“disability” OR “impairment” OR “handicap” OR “accessibility”). The search query syntax was adapted to each database to ensure consistency and accuracy.

All the results from ScienceDirect, PsycArticle, and PubMed were reviewed, and the first 200 results from Google Scholar were reviewed, adhering to the methodology outlined by Bramer et al. (2017) [24]. All the retrieved articles were imported into Zotero to facilitate screening and eliminate duplicates.

We followed the PRISMA guidelines for systematic reviews. The selection process consisted of two main stages: (1) title and abstract screening, and (2) full-text review to ensure relevance.

To be eligible for inclusion in this review, conference proceedings and journal articles had to meet stringent criteria as follows: (1) they must be written in English, (2) have undergone peer review, and (3) specifically deal with AVs and PLWD.

The initial search yielded 491 articles across all the databases. After removing 47 duplicates, 367 records were excluded based on title and abstract screening. Additionally, 28 full-text articles were excluded for not meeting the eligibility criteria. Ultimately, 51 articles were included in the core review. An additional 15 articles were identified through citation tracking. A total of 66 articles were included in the present review of the literature (see Figure 1). The comprehensive results are presented and summarized in

Table A1. Description and main results of the publications reviewed.

Authors	Country	Theme	Design	Impairment	AV Type	Nb Participant	Main Results
Agriesti et al. (2020) [57]	Finland	AVs’ impact on mobility and urban environment	Scenario-based approach	Not specified	Shared and private AVs	Not applicable	A positive influence of shared and private AV deployment on PLWD mobility is foreseen.
Alessandrini et al. (2015) [25]	Italy	AV opportunities	Position paper	Not specified	Not specified	Not applicable	AVs would allow PLWD to use cars by removing the need to drive, and increase their comfort with smoother riding.
Azizi Soldouz et al. (2020) [68]	Canada	PLWD’s views about AVs	Web-based survey	Visual impairment	Connected AVs	352	AVs are viewed to improve the independence of traveling by 78.30% of participants. Willingness to use and trust are affected by sociodemographic factors, mobility habits, past experiences, and concerns related to AVs.
Bennett et al. (2019) [75]	UK	PLWD’s views about AVs	Interview	Ambulatory impairment	Driverless vehicles	400 with ambulatory impairment 353 without impairment	Most frequent topics associated with AVs are different for participants without impairment (driver behavior (44%); unattractiveness (18%); software and control problems (21%); futuristic (17%)) and participants with ambulatory impairment (AVs are dangerous (30%); ambivalence (34%); AVs are helpful (36%)). Positive views about AVs positively influence willingness to travel in an AV.
Bennett et al. (2019) [76]	UK	PLWD’s views about AVs	Interview	Intellectual impairment	Driverless vehicles	177	Most frequent topic associated with AVs: freedom (46%); fear (34%); curiosity (20%). Positive views about AVs positively influence willingness to travel in an AV.
Bennett et al. (2020) [77]	UK	PLWD’s views about AVs	Interview	Visual impairment	Driverless vehicles	211	Most frequent topic associated with AVs: hope (37%); skepticism (24%); safety concerns (21%); affordability of AVs (18%). Positive views about AVs positively influence willingness to travel in an AV.
Bennett and Vijaygopal, (2024) [64]	UK	Mobility and transportation technology needs of people with disabilities	Imagination workshop	People with ambulatory disabilities	Autonomous vehicles	661 people with ambulatory disabilities	Self-reliant users favor assistive technologies that enhance independence, while older individuals with lower technology interest prefer familiar solutions. The study highlights the need for targeted marketing strategies and policy adaptations to address diverse mobility needs, emphasizing the growing demand for AV-integrated accessibility features and assistive technologies.
Brewer and Kameswaran, (2018) [86]	USA	AV HMI	Design-based focus group	Visual impairment	Level 3 and Level 4 AVs	15	Design sessions to create an audio and a tactile artifact aimed at helping visually impaired drivers avoid an obstacle in the case of a controlled transition. Expressed interest in being able to interact naturally with the vehicle (i.e., talking to the vehicle naturally as if it were a human).
Brinkley et al. [40]	USA	PLWD’s views about AVs	Online survey	Visual impairment	Self-driving vehicles	516	Potential benefits are more frequently evaluated as likely to occur: reduction of crashes, reduced crash severity, improved fuel economy. Most frequent concerns regarding AVs: equipment failure, system confusion in unexpected situations, interaction with pedestrians.
Brinkley et al. (2019) [90]	USA	AV HMI	Open road evaluation of an HMI	Visual impairment	Level 5 AV private or shared	20	Successful test of an interface designed to address potential issues relating to safety, cost, reliability, navigation, and accident response.
Brinkley et al. (2020) [38]	USA	PLWD’s views about AVs	Survey and focus groups	Blind visual impairment	Partially and fully autonomous vehicles	516 (survey), 38 (focus groups)	Participants were generally positive about AVs but had concerns about accessibility, trust, and user interface. Suggested improvements in navigation and feedback systems to enhance usability and independence.
Brinkley et al. (2018) [90]	USA	PLWD’s views about AVs	Focus group	Visual impairment	Self-driving vehicles	38	AVs identified as a way of improving independence and saving time. Concerns about: safety, reliability, affordability of purchase and repair, interaction with conventional vehicles, not being considered during AV development, giving AVs parking guidance, orienting at the destination, finding AVs on return in a crowded space, verifying the arrival at the correct location, not having situation awareness during operation, seeking help in case of an accident, legal liability in case of an accident, and regulation preventing AVs from operating. Preferred interaction modalities with AVs: speech interaction and with smartphones.
Brinkley et al. (2022) [91]	USA	AVs interaction	Literature review	Various disabilities, including mobility, visual, and cognitive impairments		Not applicable	The study highlights major accessibility barriers in current autonomous vehicle interfaces, such as reliance on touchscreens, lack of multimodal interactions, and inadequate voice-based controls. It emphasizes the need for inclusive HMIs that incorporate voice commands, haptic feedback, and gesture-based controls to better accommodate users with diverse disabilities.
Castillo et al. (2014) [89]	Brazil	AV HMI	Laboratory test of an algorithm	Physical impairment	“Unmanned autonomous car with a high level of autonomy”	6 with physical impairment 13 without physical impairment	Development of a brain computer interface aiming at allowing AV users with physical impairment to choose a vehicle destination.
Colley et al. (2019) [102]	Germany	eHMI	Literature review	Not specified	Not specified	Not applicable	Most of the publications about eHMI include only visual information, which could hinder communication with visually impaired pedestrians.
Colley et al. (2020) [105]	Germany	eHMI	Virtual reality and interviews	Visual impairment	Not specified	6	Test of eHMI conveying auditory information or tactile information via smartphone. Preferences towards explicit messages from AV, spoken texts, and standardized messages across manufacturers.
Cordts et al. (2021) [51]	USA	Accessibility needs and use preferences	Survey	Ambulatory, self-care, independent living, cognitive, hearing, and visual impairments	“Vehicles that can drive themselves without a human driver”	468	Most requested accessibility technologies: electronic display (49.4%); navigation app (45.3%); ramp/wheelchair access (40.4%). AVs most likely to be used as: paratransit (52.6%); personalized transport (48.1%); ridesharing service (32.5%).
Costa et al. (2018) [37]	Portugal	Universal design and automation in vehicles	Scoping review	Not specified	From Level 0 to Level 5	Not applicable	AVs from Level 4 would improve PLWD mobility.
Deka & Brown, (2021) [69]	USA	Safety perception of AVs	Telephone survey	Users of mobility devices	Fully automated vehicles	7.6% with ambulatory impairment among 1001 respondents	General population expect AVs to help people with ambulatory impairment, while people with ambulatory impairment expect AVs to decrease their safety as pedestrians.
Dicianno et al. (2021) [23]	USA	AVs and accessibility for PLWD	Literature review	Various disabilities, including visual, cognitive, hearing, and mobility impairments	Automated vehicles	Not applicable	The research identified gaps in AV accessibility for people with disabilities and emphasized the need for universal design. It recommended conducting participatory studies and adjusting policies to improve usability.
Emory et al., (2022) [58]	USA	AV policy	Literature review	Not specified	Shared and non-shared AVs	Not applicable	Literature review of policies aiming at equity in AV deployment, implying PLWD relates to vehicles’ physical accessibility and specific pricing.
Epting, (2021) [48]	USA	Moral significance of AVs	Literature review	Not specified	Automated buses	Not applicable	In some cases, it would be beneficial to keep a driver or a service agent on board to provide care to PLWD.
Etminani-Ghasrodashti et al. (2021) [53]	USA	PLWD’s view about AVs	Focus group	Physical impairment Visual impairment	Shared shuttles	20 without impairment 4 with physical or visual impairment	Accessibility equipment requested: accessible payment system, information on the booking app about the exact pick-up point and its accessibility, mat on the floor to prevent guide dogs from sliding, option to adjust the pick-up point to accommodate needs, accessible built environment, and trained operators.
Ferati et al. (2018) [81]	Norway	AV HMI	Position paper	Not specified	Not specified	Not applicable	Advocate for the use of universal design principles for the design of in-vehicle interaction.
Fink et al. (2021) [61]	USA	AV policy, accessibility, and future directions	Literature review	Visual impairment	Fully autonomous vehicles	Not applicable	Current AV policies fail to adequately address the accessibility needs of blind and visually impaired users, increasing the risk of exclusion due to discriminatory regulations. The study recommends strengthening protections under the Americans with Disabilities Act, implementing universal design principles, and integrating smartphone-based multimodal interactions to enhance accessibility and user autonomy.
Fink, Alsamsam et al. (2023) [62]	USA	User needs in fully autonomous ridesharing	Survey	Visual impairment	Fully Autonomous Vehicles	187 participants with visual impairment	Users prefer less social interaction in fully autonomous vehicles compared to human-operated rideshares, but they still want to collaborate and provide input during the ride. Key needs include real-time updates on route progress, vehicle behavior, and destination information, with a strong preference for natural language interfaces over haptic feedback.
Fink, Doore et al. (2023) [63]	USA	User-driven design of the Autonomous Vehicle Assistant to enable you to find your way around and get on board.	Survey and user interviews	Visual impairment	Fully Autonomous Vehicles	90 participants included in the survey, and 14 blind and visually impaired individuals participated in the interviews.	The Autonomous Vehicle Assistant could be a good solution to allow people with visual impairment to find their way around, and get on board, and improve their autonomy.
Goggin (2019) [46]	Australia	Communication surrounding AVs	Communication analysis	Not specified	Not specified	Not applicable	AVs associated with PLWD have been used in communication about AV development.
Golbabaei et al. (2020) [72]	Australia	AV acceptance and intention to use	Systematic review of the literature	Mobility impairment	Fully autonomous vehicles	Not applicable	The intention to use AVs is higher for PLWD.
Golbabaei et al. (2024) [45]	Australia	Designing and developing accessible autonomous vehicles	Systematic review of the literature	Not specified	Not specified	Not applicable	AVs could significantly improve mobility and inclusion for PLWD by enhancing access to work, healthcare, and social activities. The study recommends universal design principles, multimodal interaction (audio, tactile, visual), financial subsidies, and specialized AVs to ensure equitable transport solutions.
Goralzik et al. (2022) [32]	Europe (21 countries)	Accessibility assessment of shared mobility services	Survey	Visual impairment Mobility impairment Multiple disabilities	Shared mobility services (ride pooling, microtransit, robotaxis, motorbike taxis, e-scooter sharing, bike sharing)	553 individuals with disabilities	Microtransit, robotaxis, and ride pooling had the highest accessibility ratings, while motorbike taxis, e-scooter sharing, and bike sharing were seen as least accessible. None of the shared mobility services fully met the access needs of disabled users in their current form.
Guerrero-Ibañez et al. (2023) [67]	Spain	Assistive self-driving car networks for disabled road users, developing a model using deep learning and wireless communication	Descriptive study with an experimental design	Visual impairment Hearing impairment Mobility impairment	Self-driving vehicles	Not applicable	Proposed a framework integrating assistive technology into self-driving vehicles to improve interaction with disabled pedestrians by developing a hand gesture recognition model and a disabled user identification system using wearable devices.
Harkin et al. (2024) [66]	Germany	Perception of AVs	Focus group	People with walking disabilities	Not specified	22 participants, including 3 people with walking disabilities	AVs are positively perceived due to their rule-abiding and cautious driving behavior, but concerns expressed about the lack of explicit communication and increased risks in mixed traffic (AVs and conventional vehicles).
Harper et al. (2016) [60]	USA	Travel increase estimation	Miles traveled estimation	Adults with a “medical condition that makes it hard to travel”	Not specified	Not applicable	Estimation of an increase of 2.5% of miles traveled by adults with a medical condition making it hard to travel; their mobility increased with AVs to the level of the non-restricted population.
Hwang et al. (2020) [39]	USA	PLWD’s views about AVs	Focus group	Visual impairment Physical impairment	On-demand public transport	13 with visual impairment 11 with physical impairment 9 transit services experts	AVs perceived as a way to increase freedom of travel and decrease traveling costs. Concerns about safety, accessibility, and adverse effects.
Hwang & Kim, (2023) [47]	South Korea	Factors influencing the choice of autonomous vehicle transportation services	Survey	Visual impairment Physical impairment	Fully autonomous vehicles) for public transportation	110 with physical disabilities and 36 with visual impairment	Individuals with disabilities who had a negative experience with public transit were more likely to prefer autonomous vehicle transportation services. The presence of an onboard human attendant significantly increased trust and the likelihood of choosing autonomous vehicles.
Hwang et al. (2021) [56]	USA	PLWD’s views about AVs	Focus groups and survey	Visual impairment Physical impairment	Fully autonomous vehicles) for public transportation	150 individuals with disabilities (various physical and visual impairments) and 72 transportation professionals, policy makers, and caregivers	People with disabilities had high acceptance of AVs but emphasized accessibility, safety, and reliability concerns. They expressed frustration with current public transit services and saw AVs as a potential solution.
Jeon et al. (2016) [36]	USA	AV challenges and opportunities	Position paper	Visual, auditory, cognitive, motor impairments	Low-speed urban shuttles	Not applicable	Impact of AVs on PLWD’s mobility would differ according to the level of automation, being higher for Levels 4 and 5.
Kacperski et al. (2024) [44]	Germany	AV acceptance	Survey	Visual impairment	Connected and autonomous vehicle Level 5	114 with visual impairments and 117 without visual impairments	More positive AV attitudes in participants with visual impairments compared to those without visual impairments, mainly due to higher hopes for independence and optimistic views on safety and sustainability.
Kassens-noor et al. (2020) [74]	USA	Willingness to use AVs	Onboard intercept survey	Mobility impairment	Autonomous buses and shuttles	271 with mobility impairment among 1468 respondents	People with mobility impairment are less willing to ride AVs.
Kassens-noor et al. (2021) [65]	USA	Perception and willingness to use AVs	Onboard intercept survey	Visual impairment, Mobility impairment	Autonomous buses and shuttles	1861 public transit users in Michigan (40% with special needs)	People with special needs rely more on public transit. Visually impaired individuals are more willing to use public AVs, while those with mobility disabilities have concerns about safety and trust in automation.
Kempapidis et al. (2020) [79]	UK	User experience	On-road experiment	Visual, coordination, dexterity, hearing, speech, and mobility impairments Chronic illness Other disabilities	Automated shuttles	228 visual impairment, 3 chronic illness, 12 coordination or dexterity impairment, 16 hearing or speech difficulty, 60 mobility-related disability, 136 other disabilities, among a total of 419	Positive experience and positive emotions during a journey on board an automated shuttle.
Khan et al. (2022) [31]	USA	Planning AV introduction in current paratransit services	Modeling	PLWD eligible to paratransit	Shared AVs	Not applicable	A shared AV fleet, including accessible vehicles, could complement the actual paratransit system.
Klinich et al. (2022) [83]	USA	Securement systems for wheelchair users	Literature review	Wheelchair users	Shuttles	Not applicable	Current wheelchair security systems are not all compatible with shared AVs.
Kutela et al. (2025) [78]	USA	Impact of AVs on independence	Nationwide survey (Bayesian Network Analysis)	Not specified	Not specified	4642	AVs could enhance the independence of people with disabilities. Concerns include accessibility, system reliability, and affordability, which impact willingness to adopt AVs. The study recommends inclusive design, financial incentives, and policy measures to ensure AV accessibility for PLWD.
Kuzio, (2021) [27]	USA	Legal framework surrounding AV use by PLWD	Legal analysis	Not specified	Not specified	Not applicable	AVs could increase PLWD’s mobility as they can be ridden like a regular car without the need to be able to drive and could make paratransit more flexible and cheaper.
Mandujano Nava et al. (2017) [87]	Mexico	AV interior design	Design concept creation	Wheelchair users	“Modular mobile vehicle, autonomous and manually operated”	Not applicable	Description of the process of designing the concept of AVs allowing the boarding of wheelchair users.
Martin, (2018) [29]	USA	AVs’ impact on social justice and environmental sustainability	Ecosocial parameters analysis	Not specified	Not specified	Not applicable	AV could increase PLWD’s mobility because they can be ridden like a regular car without the need to be able to drive, thanks to a loosening of driving restrictions.
Miller et al. (2022) [34]	Singapore	Acceptance of shared AVs and communication needs	Survey and focused group discussions	Blindness Deafness Mobility impairments Autism	Shared AVs in public transport	300 (survey), 53 (focus groups)	People with disabilities had positive attitudes toward shared autonomous vehicles but highlighted concerns about safety, accessibility, and trust. Essential features include live intercoms, visual/auditory cues, and wheelchair-friendly access.
Ohnemus & Perl, (2016) [30]	Canada	AVs’ effect on land use	Position paper based	‘Disabled elderly’ ‘Disabled’	Shared and non-shared AVs	Not applicable	Disabled elderly would be more inclined to adopt shared AVs compared to private AVs as it would have a lower cost and risk. Shared AVs would offer an additional mobility option to PLWD unable to travel with fixed-route transit.
Papa & Ferreira, (2018) [59]	UK	Societal consequences of AVs deployment	Scenario-based approach	“People with some form of health condition, disability, and visual impairment”	Fully automated vehicles	Not applicable	PLWD could benefit from AVs and digital technology associated with AVs as they could travel without the need to operate a vehicle and would benefit from the monitoring of their health status on board.
Park et al. (2023) [54]	USA	AV interface	Phone interviews and focus group meetings	Older adults with cognitive impairment	AVs Level 5	14 with cognitive impairment and 9 caregivers	The AV interface could be customized based on the severity of cognitive impairment, use familiar terminology, include reminders, and adopt a navigation app-like design (e.g., Google Maps), with seamless page transitions for ease of use by individuals with cognitive impairments.
Patel et al. (2021) [52]	USA	PLWD’s views about AVs	Focus group	Visual impairment Physical impairment	AVs integrated into the public transportation system	3 with visual impairment 1 with physical impairment	Participants expressed accommodation needs related to the accessibility of equipment, the booking app, the built environment, and onboard assistance.
Petrovic et al. (2022) [33]	Serbia	Perspective of the driving status	Survey	Physical impairment	AVs in public transport	160 (80 drivers and 80 non-drivers with physical disabilities)	Non-drivers with physical disabilities are more likely to use autonomous vehicles than drivers. Trust, accessibility, and attitudes significantly impact AV acceptance.
Ranjbar et al. (2022) [55]	Sweden	Vibrotactile guidance for trips with AVs	Case study using Wizard-of-Oz simulated AV	Hearing Visual impairment	Simulated AV (Wizard-of-Oz)	15 (5 blind, 5 deafblind, 5 deaf participants)	Vibrotactile guidance improved independent navigation before, during, and after the trip. Participants valued technology but suggested enhancements such as different vibration intensities for critical information.
Robert, (2021) [104]	USA	Interaction with AVs	Model presentation	Not specified	AVs from Level 4	Not applicable	Need to be more inclusive in the view of human characteristics when designing interactions with AVs.
Roundtree et al. (2020) [101]	USA	eHMI	Design guidelines based on a literature review about crossing behavior	Not specified	Not specified	Not applicable	Design guidelines related to the way AV intent should be communicated, messages’ sensorial modalities, visual displays’ position, messages conveyed, and message format.
Son et al. (2019) [88]	Korea	AV HMI	AV HMI	Auditory impairment Visual impairment	“Fully autonomous vehicles”	Not applicable	Technical design of an interface meant to inform visually and auditory impaired users about vehicle self-diagnosis.
Sultan & Thomas, (2020) [70]	UK	PLWD’s views about AVs	Focus group	People with epilepsy	“Fully autonomous vehicles” operated without a driving license	8 people with epilepsy or caregivers	Concerns in case of a seizure and the stigma about being identified as epileptic AVs identified as a way to improve independence and freedom from licensing, and allowing the user to adapt more easily to treatment to reduce side effects.
Sundararajan et al. (2019) [80]	USA	Recommendations and requirements for AV accessibility	Panel discussion	Not specified	Not specified	Not applicable	Advocate for the use of universal design principles, considering the accessibility of the built environment, cooperation towards standards, policies, and regulation for accessibility.
Tabattanon & D’Souza, (2021) [82]	USA	AVs interior accessibility	Guidelines analysis Expert recommendation Accessibility testing	Mobility impairment	Automated shuttles	40	Accessibility constraints should be considered early in the design process.
Tabattanon et al. (2019) [85]	USA	AV interior accessibility	Systematic review of the literature	Mobility impairment	Automated shuttles	Not applicable	Creation of a Web-based repository gathering publications about transport accessibility to PLWD and elderly adults.
Tabattanon et al. (2021) [84]	USA	AV interior accessibility	Mock-up testing with participants	Wheelchair users	Automated shuttles	6	Presentation of a shuttle’s full-scale mock-up used to evaluate vehicle accessibility for wheelchair users.
Wang et al. (2021) [73]	Canada	Willingness to pay for AVs	Stated preference survey	Person with a disabled family member	Ride-hailing, shared or private vehicles with automation from 0 to 5	190	Having a disabled family member is associated with a higher willingness to pay for an AV.
Wu et al. (2021) [28]	USA	AVs’ impact on mobility	Literature review of mobility behavior	Not specified	AV of high and full automation not requiring a driver’s license	Not applicable	AVs could increase PLWD’s mobility as they could be ridden like a regular car without the need to be able to drive.
Zmud et al. (2016) [71]	USA	Intention to use AVs	Online survey	“Travel-restrictive medical condition”	Level 5 private or shared vehicles	556 respondents, of whom 2% had a travel-restrictive medical condition	Being impaired is associated with an increased intention to use AVs.

.

3. Results

3.1. Opportunities for PLWD Based on Different AV Transport Modes

3.1.1. Personal AVs

AVs present a promising opportunity for enhancing mobility for PLWD. The potential benefits encompass the prospect of PLWD using AVs as vehicles that offer the freedom of travel without the need for active driving [25,26,27,28]. The relaxation of driving restrictions is seen as a plausible outcome [29], introducing an additional mobility option for PLWD [30]. Additionally, AVs could remove barriers encountered in accessing public transport, such as infrastructure limitations, lack of real-time assistance, and complex navigation requirements [26]. Indeed, removing the cost of active driving could allow for improved flexibility (i.e., extended riding hours) or reduced travel costs [27].

3.1.2. Shared and Public Autonomous Transport

The benefits mentioned above relate to the use of AVs as private vehicles; however, public services such as paratransit (i.e., transportation services provided to PLWD in the USA) could be improved. Indeed, eliminating the cost of active driving could enable greater flexibility (i.e., extended service schedules) or reduced travel costs [27]. In addition, paratransit drivers would be freed from the task of driving, allowing them to concentrate on picking up passengers [28]. In this case, the advantages previously highlighted by Kuzio (2021) [27] might be diminished. In addition, a shared AV fleet incorporating accessible vehicles could improve transportation systems by complementing existing paratransit services [31].

Moreover, the prospect of private AVs could redefine the role of paratransit services, freeing drivers from the driving task to focus on passenger care [28]. Incorporating accessible vehicles into shared AV fleets represents another avenue for improving transit services for PLWD. This integration not only enhances existing paratransit services but also contributes to the overall improvement of transit systems [31]. In this regard, robotaxis are increasingly viewed as a promising solution to improve accessibility for PLWD [32]. The economic advantage of shared AVs, coupled with the potential for improved comfort due to their smoother control [25], makes them an attractive option, particularly for elderly PLWD [30].

Private AVs enhance autonomy but may reinforce socio-economic disparities due to affordability challenges [33]. Shared AVs, while more accessible, require substantial infrastructure improvements to ensure seamless usability for individuals using mobility aids [34]. Optimizing both models to balance affordability, accessibility, and independence remains crucial.

3.1.3. Influence of Automation Levels on User Expectations

Initially, an analysis was conducted on how AVs could impact the accessibility of transport-disadvantaged groups, including people with physical and sensory impairments [35]. Different impacts on the accessibility of AVs for PLWD according to their level of automation and mobility service models (i.e., private or shared vehicles) are foreseen. Specifically, when considering private AVs up to Level 3, no changes are expected as “a fully capable driver will still be needed to take over control of the vehicle” (p. 68). However, for private or shared AVs from Levels 4 to 5, it is foreseen that there will either be no change or a possible opportunity to improve access, but this will be associated with potential barriers due to higher purchase or operating costs caused by a custom design to accommodate PLWD’s needs. This perspective on the differing impacts of AVs based on the level of automation (i.e., Level 3 vs. Levels 4 and 5) is shared by Jeon et al. (2016) [36] and Costa et al. (2018) [37]. While fully autonomous vehicles (Level 5) generate high expectations, they also raise safety concerns. Level 4 automation, which retains human intervention capabilities, is generally preferred for reliability [38].

3.2. Expected Benefits and Key Concerns

3.2.1. Anticipated Benefits

The potential positive influence of AVs on mobility reported by Hwang et al. (2020) [39] was also acknowledged by visually impaired participants [40]. A comprehensive survey study involving 516 visually impaired respondents in the United States examined their acceptance, concerns, and willingness to purchase partially and fully automated vehicles. Additionally, a qualitative study exploring the perspectives of 38 blind and low-vision participants using a focus group methodology provided further insight into the perceptions of AV technology. The findings suggest that visually impaired individuals express optimism about AVs’ potential to enhance mobility and independence [38]. They identified AVs as a means to enhance independence by eliminating reliance on others for their mobility needs. Furthermore, AVs are seen as a time-saving tool, addressing the existing barriers in the current mobility system. Indeed, waiting time when using paratransit is often cited as a disadvantage (e.g., [8]).

Furthermore, this technology has the potential to lower transportation costs, making mobility more affordable [41]. Moreover, the development of car-sharing models with AVs can reduce the need for personal vehicle ownership, thereby lowering individual transportation costs. This is particularly beneficial in urban settings where parking and maintenance costs are high [25]. AVs can enhance traffic flow by maintaining optimal speeds and safe distances, which increases outflow and reduces stop-and-go traffic, especially at higher penetration rates [42]. However, they can also increase travel demand and vehicle miles traveled, potentially leading to more congestion if not managed properly [43]. Beyond these functional benefits, AVs promote greater social participation by enabling non-drivers with disabilities to engage more actively in economic and community life [44,45].

A mixed-methods study examined the perceptions of individuals with disabilities regarding AV transportation services, incorporating focus groups, surveys, and stated preference experiments. In this study, 222 participants from Austin and Houston found that 74% of respondents either strongly agreed or agreed that AV transportation could help address the transportation challenges faced by PLWD. This interest in AV adoption was largely driven by frustrations with public transit inefficiencies, as well as the inaccessibility of existing transportation options.

Nevertheless, even if AVs as a potential opportunity for PLWD have been used as a topic of communication around AV development [46], questions and potential barriers related to their use by PLWD are raised. Concerns about safety, reliability, and the absence of human attendants remain significant. As reliable information about AV technology continues to spread, anxieties regarding mechanical failures and safety risks may decline [47]. However, in a study involving 146 individuals with disabilities (75% with physical disabilities and 25% with visual impairments), the majority expressed anxiety over traveling unassisted in AVs, highlighting the need for human attendants to enhance user confidence and accessibility [47].

The necessity of an onboard operator, as emphasized by PLWD, is also supported by transit service experts. Wu et al. (2021) [28] explored the evolving role of “drivers”, suggesting a shift from traditional driving responsibilities to customer service-oriented roles. Similarly, Epting (2021) [48] underscored the critical role of human operators in maintaining service quality for PLWD. To ensure an inclusive and equitable AV transportation system, future deployments should consider maintaining human support roles where needed to enhance accessibility and user confidence.

3.2.2. Main Concerns

Expectations and concerns regarding AV use among PLWD vary based on disability type (e.g., visual versus physical impairment) and mobility expertise. This underscores the importance of engaging diverse populations in research to ensure comprehensive and inclusive insights [39].

While many PLWD perceive AVs as safer than human-driven vehicles, key concerns remain. These include the absence of a driver, interactions with other passengers, emergency situations, and system malfunctions. Transit service experts also emphasize the necessity of accessible pick-up/drop-off areas and built-in accessibility features. Furthermore, PLWD express concerns regarding communication with AVs, highlighting the importance of offering multiple interaction modalities tailored to diverse needs.

The potential unintended consequences of AV deployment include increased congestion due to heightened travel demand and potential job losses resulting from driver automation. While some PLWD anticipate lower service costs due to reduced labor expenses, transit experts predict that cost reductions may only materialize in the long term due to ongoing technical development and infrastructure adaptations [39].

Certain concerns are shared across all user groups, such as safety, reliability, and affordability, while others are specific to visually impaired individuals. These include insufficient consideration of their needs in AV design, challenges in providing navigation or parking guidance, difficulties in locating the vehicle, verifying the correct destination, maintaining situational awareness during transit, and seeking assistance in case of accidents or emergencies. Legal concerns also persist, particularly regarding requirements for a valid driver’s license for AV operation [27,29,49,50].

Participants express a strong preference for speech interaction with AVs and smartphone-based controls. Additionally, Cordts et al. (2021) [51] highlight varied preferences for AV service models, with 52.6% of respondents envisioning AV use in paratransit, 48.1% in personalized transport, and 32.5% in ridesharing services. While most participants rate AV safety favorably, 32.3% report feeling safer with an onboard driver.

A study by Patel et al. (2021) [52] focusing on individuals with physical and visual impairments identified essential accommodations, including AV access from healthcare facilities, accessible pick-up/drop-off points, wheelchair accessibility, onboard assistance, a user-friendly booking app with integrated payment, and guide-dog-friendly vehicle interiors. These findings emphasize the importance of inclusive AV design.

Accessibility concerns differ significantly between PLWD and the general population. Unique requests from PLWD include an accessible payment system, clear pick-up point details, floor mats for guide dogs, flexible pick-up adjustments, and trained AV operators [53]. These distinctions highlight the need for AV services that prioritize universal accessibility.

Older adults with mild to moderate cognitive impairments express significant concerns about AV trust, safety, and the lack of human assistance, with memory-related difficulties making navigation particularly challenging [54]. In a Serbian study of 160 individuals with physical disabilities (80 drivers, 80 non-drivers), skepticism about AV safety and reliability remained prominent among drivers [33]. Safety concerns also emerge among individuals with blindness, deaf-blindness, and deafness [55]. A mixed-methods study examining AV perceptions among 222 PLWD in Austin and Houston found that safety, reliability, and lack of human assistance remain major barriers, with half of the respondents preferring an onboard human operator [56]. Mode choice experiments indicate a strong preference for AV transportation services (61.6%) over personal vehicles, paratransit, or buses, although most favor single-ride AVs over shared options, raising concerns about cost and efficiency.

3.3. Legal and Regulatory Considerations

3.3.1. Licensing and Regulatory Barriers

Licensing policies pose significant barriers to AV adoption for PLWD. Hancock et al. (2020) [26] pose critical questions about how to license and train PLWD for AV use, raising concerns about adapting the level of vehicle automation to meet their needs. Licensing issues, rooted in conventional definitions that view a driver as a “person” capable of controlling the vehicle, present a significant barrier [29,49]. The Vienna Convention, aimed at harmonizing road regulations, necessitates the driver’s ability to control the vehicle, potentially excluding non-licensed PLWD from AV use [49]. Proposed solutions, such as granting “legal personhood” to software drivers, introduce ethical and responsibility considerations, particularly in the event of accidents [49]. An alternative suggestion involves redefining the term “driver” in the Vienna Convention to encompass both a “person or a thing” (e.g., software). In this definition, PLWD would be able to travel in the vehicle without having to be considered as the driver, but rather as the vehicle’s passenger. Francis (2018) [50] addressed the licensing issue and argues that even fully automated vehicles would require licensing to ensure the user knows where the vehicle can be instructed to drive. However, failure to provide accommodation in the licensing rules to make them accessible to PLWD would represent discrimination. The question of licensing for private AVs is still an issue, even if solutions are identified, so that AVs would not become a cause of discrimination against PLWD.

3.3.2. Employment and Work Accessibility

Francis (2018) [50] analyzed other legal issues arising from the use of AVs by PLWD, particularly the impact of driving restrictions on their employment opportunities. In the United States, the Americans with Disabilities Act prohibits employment discrimination based on disability and requires employers to provide “reasonable accommodations” to enable PLWD to perform their jobs. However, for jobs requiring operating a vehicle as an “essential function of the position”, employees might be disqualified due to their inability to drive. The use of AVs could allow them to perform job functions that involve vehicle operation, even if they are unable to drive a conventional vehicle.

3.3.3. Privacy and Data Security

The question related to data privacy, particularly regarding the sharing of medical data to address the needs of PLWD, has also been raised by Francis (2018) [50]. This legal issue was likewise noted by Kuzio (2021) [27], who argued that it might be requested when using shared AVs to provide data, and that PLWD have the additional risk of having their medical and health data hacked. Privacy concerns have been shared, with fears of increased surveillance and data misuse also recently noted in a study carried out in Germany [44]. In this study, over 65% of visually impaired participants expected their privacy to deteriorate with the introduction of AVs, compared with 77% of non-impaired participants. The analyses revealed that visually impaired people expressed less concern about the impact of AVs on personal data privacy, although this concern remained significant. These findings are part of wider discussions on the accessibility and regulation of AVs. Kuzio (2021) [27] highlights the need for a robust legal framework to ensure AV accessibility for PLWD and proposes to “expand Americans with Disabilities Act’s protective framework to include AVs”, “establish Standards for Accessible AVs”, “encourage Pilot Testing of AVs for Paratransit”, and provide “guidance at the Federal Level for Transit Agencies”.

3.3.4. Policy and Regulatory Scenarios

Beyond the legal considerations, scenarios and critical decisions allowing AVs to serve as an opportunity for the mobility of PLWD were also explored [35,36,37,57,58]. Policies will play a crucial role in ensuring AV accessibility for PLWD. Papa and Ferreira (2018) [59] used a scenario-based approach to identify critical decisions (i.e., “policy choices related to the governance process of developing and implementing AVs”, p. 4) that could positively or negatively influence AV accessibility. The authors examined several scenarios considering the potential impact of AVs on well-being and the environmental and social tensions (e.g., aggravation of social inequalities). Interestingly, only the optimistic scenario addressed PLWD’s mobility, with this element not being addressed in the pessimistic scenario of the integration of AVs in our daily life. According to the authors, PLWD would benefit from AVs, as this would allow them to travel without the need to operate a vehicle, and to benefit from the digital technology associated with AVs via the onboard monitoring of their health status, allowing them priority in traffic to reach medical assistance.

The scenario-based approach was also used to assess the impact of AV development on public transportation, parking, social aspects, pedestrian and human safety, and density and sprawl with a focus on the urban environment [57]. The three scenarios examined considered either the limited development of AVs or the development of AVs mainly as private or shared vehicles. The identified social impact of AVs included a positive effect for PLWD, as AVs would give them the opportunity to be included in mixed mobility. However, the authors do not shed light on the difference between the shared and private AV scenarios. Furthermore, policies also play a pivotal role in shaping opportunities for PLWD in terms of mobility. Emory et al. (2022) [58] highlighted that policies aiming to ensure a specific number of AVs in a shared fleet are accessible, and coupled with tailored pricing for PLWD, could significantly impact AV development. Additionally, Harper et al. (2016) [60] delved into the potential impact of policies on increasing mobility for PLWD. They estimated a future surge in vehicle miles traveled facilitated by AVs, focusing on currently underserved populations such as adult non-drivers, individuals aged over 65 without medical conditions, and adults with travel-restrictive medical conditions. A national survey on transportation in a fully AV environment estimated that AVs could increase the total vehicle miles traveled by adults with travel-restrictive medical conditions by 2.5%. This increase is lower than that estimated for non-driver adults (not considering restrictive medical conditions) (9%) but higher than that estimated for elderly drivers without a medical condition (2.2%). It is important to note that this study considered “adults with a medical condition that make it hard to travel” (p. 4), and thus, some PLWD may not have been considered in these estimations based on this definition.

3.4. Social Participation and Inclusion

3.4.1. Enhancing Mobility and Independence

Enhancing mobility opportunities for PLWD through AVs involves understanding the potential barriers and perspectives of PLWD regarding AVs. This understanding is crucial for developing guidelines aimed at fostering AV accessibility and adoption. While studying the views of PLWD is fundamental, alternative methods are available for deriving guidelines to design accessible AVs. Fink et al. (2021) [61] demonstrate that AVs significantly improve autonomy for PLWD but require policy interventions to ensure equitable access. AV adoption may also depend on the integration of smartphone-based accessibility tools [62]. Fink, Doore et al. (2023) [63] identify key barriers for visually impaired users, including precise vehicle localization and safe pedestrian–vehicle interactions. Their study finds that ultra-wideband guidance systems significantly enhance navigation precision, achieving a 90% success rate in vehicle localization tests.

AVs have the potential to improve quality of life for PLWD by increasing travel autonomy and reducing dependency on caregivers [61,64].

3.4.2. Trust and Public Perception

Despite growing interest in AVs among PLWD, concerns over reliability, safety, and lack of human assistance persist [65]. Kassens-Noor et al. (2021) [65] report that 56% of individuals with disabilities hold negative perceptions of AVs, particularly those with mobility impairments. A study involving people with walking difficulties revealed concerns about the lack of communication with the vehicle (due to the absence of a driver) and increased risks in mixed traffic conditions (with both AVs and conventional vehicles) [66]. Hwang and Kim (2023) [47] indicate that trust in AVs increases as information about technological advancements spreads, reducing anxiety about mechanical failures. However, Francis (2018) [50] raises concerns regarding trust in AVs related to data privacy, particularly with the sharing of medical data to address the needs of PLWD. Guerrero-Ibañez et al. (2023) [67] propose that advanced pedestrian detection and deep learning-based interaction systems can enhance user confidence in AVs by improving safety features and predictability. Incorporating user-friendly interfaces, real-time feedback, and transparent safety measures is essential for fostering public trust and promoting the broader adoption of AV technology among people with disabilities. Trust-building measures, such as onboard human assistance and reliable safety mechanisms, are crucial for adoption.

3.5. Factors Influencing Adoption and Willingness to Use AVs

3.5.1. General Impressions of AVs Among PLWD

PLWD perceive AVs as a promising opportunity to improve their mobility. Azizi Soldouz et al. (2020) [68] surveyed 352 visually impaired participants, revealing that 78.30% of them considered AVs as a way of improving their independence in traveling. Similarly, a study conducted by Brinkley, Gilbert et al. (2018) [40] involving 516 visually impaired participants reported a large majority of positive opinions about AVs, with 50.18% expressing extreme positivity, 30.44% moderate positivity, and 7.75% slight positivity. Those participants also evaluated potential benefits and concerns about AVs. Among the eight benefits considered, the three most likely to occur were a reduction in crashes, reduced severity of crashes, and improved fuel economy. Among the 10 concerns regarding AVs, the 3 most frequently cited were equipment failure, system confusion in unexpected situations, and system interaction with pedestrians [40].

Despite these concerns, 93.31% of the participants expressed interest in owning an AV. However, the AV benefits and concerns evaluated in this study were not specific to PLWD (i.e., foreseen to be specifically encountered by PLWD). Moreover, as this study did not include a control group, it does not allow for a comparison of PLWD’s views about AVs and those of the rest of the population. Moreover, people with higher levels of educational attainment (e.g., bachelor’s degree or more) have concerns about whether their needs are being adequately considered in the design of this emerging technology [38].

3.5.2. Pedestrian Interaction and Safety Considerations

Concerns about interaction with pedestrians mirror the concerns mentioned by Deka and Brown (2021) [69]. While the general population expects AVs to benefit people with ambulatory impairments, those with such impairments anticipate a decrease in their safety as pedestrians. Additionally, a focus group exploring the view of people with epilepsy regarding AVs was also conducted by Sultan and Thomas (2020) [70] and unveiled a positive outlook on AVs, considering them instrumental in enhancing independence and managing the challenges of the condition. The freedom from licensing constraints offered by AVs was seen as facilitating easier adaptation and reducing side effects.

3.5.3. Variability in Willingness to Use AVs Depending on Disability and Profile

Several studies have delved into PLWD’s attitudes and willingness to use AVs. Zmud et al. (2016) [71], cited by Golbabaei et al. (2020) [72], reported an increased intention to use AVs by PLWD (those with a physical condition preventing driving) compared to others, while Wang et al. (2021) [73] showed higher willingness to pay for AVs by those with a family member living with a disability. Conversely, Kassens-Noor et al. (2020) [74] reported a lower willingness among PLWD to ride AVs. This disparity underscores the influence of factors such as the specific population interviewed (e.g., PLWD versus family member) and the metric being assessed (e.g., willingness to pay versus willingness to ride).

In a series of three experiments, Bennett et al. [75,76,77] explored how individuals with disabilities (ambulatory impairment [75], intellectual impairment [76], and visual impairment [77]) perceived AVs compared to a control group. The results indicated that perception varied significantly by type of disability. People with ambulatory impairments highlighted dangerous aspects and helpful features, but remained ambivalent. Individuals with intellectual impairment expressed fear, curiosity, and a sense of freedom. Finally, those with visual impairments conveyed hope, skepticism, affordability concerns, and safety worries. These findings highlight the need to adapt AV design and policy initiatives to the specific populations of people living with disabilities. In a related study involving 352 participants with visual impairment, Azizi Soldouz et al., 2020 [68] uncovered additional factors influencing the willingness to use AVs and the level of trust associated with them. Sociodemographic factors (e.g., being blind or partially sighted from birth, being a woman, not having a college degree, being over 65 years old), mobility habits, past experiences (e.g., being satisfied with public or adapted transport, having experienced a near-accident, using mobile applications and technological devices for navigation), and concerns related to AVs (e.g., concerns about communication with AVs, vehicle moral choices, technical failures, low-noise issues) emerged as significant determinants. Visually impaired individuals were generally more willing to use AV technology, viewing it as a tool for greater independence. In contrast, individuals with mobility restrictions exhibited lower willingness, citing concerns about safety, accessibility, and technological reliability [65].

Support for driverless passenger vehicles for the elderly and PLWD is influenced by broader sociodemographic factors [78]. Higher income correlates with greater support, as respondents earning more than USD 100,000 are 12.7% more likely to approve of such vehicles than those earning less than USD 30,000. Education also plays a crucial role, with those with a university degree or postgraduate diploma showing a level of support almost twice as high as those with a university degree or some university experience, compared to those with a high school diploma or less. This study, conducted in the United States, shows that political affiliation also has an impact on opinions about AVs. Republicans are less likely to support AVs than Democrats, with a predicted probability difference of around 6.5%. These results indicate that public support for autonomous mobility solutions is shaped by a combination of many factors (e.g., socio-economic and political), reflecting the complexity of attitudes towards AV technology. Finally, gender and location impact interest in AVs [33]. In a study among persons with physical disabilities in Serbia using an online questionnaire completed by 160 respondents, Petrovic et al. (2022) [33] showed that urban residents were more open to adopting AVs, and that women, especially those who did not drive, showed greater interest in AVs.

3.5.4. Impact of First Interaction with AVs

Beyond willingness to use AVs, some studies have examined initial experiences and real-world interactions with AV technology. Kempapidis et al. (2020) [79] studied the first interactions with an automated shuttle among 419 PLWD, who tested the vehicle in a closed area during a 7-min journey. The evaluation based on questionnaires and facial emotion analysis revealed a positive experience, with happiness and surprise being the predominant emotions expressed. As such, no specific barriers to using the shuttle were highlighted. However, a deeper inquiry, with interviews, for instance, could give further insight into the participants’ experience. Additionally, this study was conducted within a specific population (i.e., veterans) and a controlled environment, limiting the generalizability of its findings.

3.6. Design Recommendations for Accessible AVs

3.6.1. Improving Vehicle Accessibility

Principles of universal design. Panel discussions provide general recommendations and requirements for AV accessibility for PLWD [80]. They advocate the application of universal design principles and emphasize considering the accessibility of the built environment and the need to share data about its accessibility features. Additionally, they highlight the importance of cooperation among stakeholders to standardize and share data, aiming to establish “updated standards, policies and regulatory frameworks for universal accessibility” (p. 89). The use of universal design principles was also advocated by Ferati et al. (2018) [81] for the design of AV in-vehicle interaction. They also recommended conveying critical pieces of safety information through a modality that will not be customizable by the driver and chosen via an empirical process. Furthermore, they suggested allowing customization, such as the option to remove non-critical safety messages.

Physical accessibility for users with mobility impairments. The importance of the early consideration of PLWD needs has been underscored by Tabattanon and D’Souza (2021) [82]. They retrofitted a commercially available automated shuttle to meet the conditions outlined in the Americans with Disabilities Act Accessibility Guidelines for people with mobility impairments. Accessibility specialists identified inadequacies of the vehicles related to the boarding of mobility-impaired passengers, such as the absence of a wheelchair securement system, insufficient clear floor space and turn space, or a level change at the door. An interactive tool and accessibility standards were employed to recommend modifications, including the installation of an automated access ramp, an increase in the available floor space, and the installation of a securement system for wheeled mobility devices. Implementing these recommendations led to modifications in various shuttle components, compromising other vehicle characteristics, such as energy efficiency. However, despite compliance with minimum requirements, the authors noted that the vehicle dimensions “were less than the recommended dimensions for maximizing inclusion of different mobility device types” (p. 3). The interior vehicle design was tested with 40 mobility-impaired participants, revealing challenges in fully meeting accessibility needs due to the constraints of the original design. Thus, the authors recommended initiating discussions about accessibility earlier in the design process. A literature review on AV accessibility for wheelchair users was undertaken by Klinich et al. (2022) [83]. The review focused on securement systems available to wheelchair users, revealing that the current available system may not be deployable in future AVs, especially in shared vehicles. Most security systems from individual modified vehicles may not be adaptable to shared AV fleets, as they are specific to each driver. Additionally, security systems in current public transport vehicles are not suitable for smaller vehicles such as shuttles, due to safety reasons. The literature review emphasized the need for harmonization efforts to ensure compatibility between wheelchairs and shuttles. Proposing a methodology to generate design guidelines for the interior space of automated shuttles, Tabattanon et al. (2021) [84] conducted two experiments with wheelchair users using a full-scale mock-up of a shuttle. The mock-up included adjustable cabin dimensions, an adaptable folding access ramp, door position variations, grab-bars, movable seats, and an electronic semi-automated wheeled mobility securement system. The experiments aimed to study the influence of vehicle parameters such as available floor space and doorway location on ingress and egress times for wheelchair users. The preliminary results from six participants suggested a preference for a larger available floor space, although no significant effect of vehicle parameters was reported. Despite this, the methodology holds promise for providing precise design recommendations for automated shuttle design in the future.

Accessibility for people with sensory impairments. Fink, Doore et al. (2023) [63] studied the accessibility problems encountered by visually impaired users, highlighting difficulties related to accurate vehicle location, obstacle avoidance, and the absence of human assistance. They evaluated multimodal assistance solutions, precise guidance based on ultra-wideband (UWB) technology, and computer vision for obstacle detection. In a trial with 10 blindfolded participants, GPS alone proved ineffective (0% success rate), with location errors of up to 21.9 m. In contrast, the combination of GPS and UWB achieved a 90% success rate, significantly improving navigation accuracy and enabling participants to accurately locate the vehicle door. Further tests with six visually impaired people confirmed these results: all the participants successfully avoided obstacles, and all were able to locate the vehicle door and handle thanks to the audio and haptic guidance. Users found the application useful and accessible, appreciated the customization of the user profile and suggested improvements, such as adding several emergency contacts and integrating autonomous vehicle assistants with existing transport services such as Uber and Lyft.

Personalization and adaptive interfaces. Ferati et al. (2018) [81] proposed that vehicle design should allow users to personalize or customize their experience. This approach would allow users to select the interface modality that is most convenient or accessible to them. The implementation of user profiles has been suggested to avoid users having to provide interface preferences each time they access the vehicle. Instead, these systems could utilize devices such as the users’ smartphones, wearables, or Internet of Things technologies to store interface preferences. These types of systems could automatically adapt the vehicle interface to specific users’ needs. However, safety should be considered over user preferences for information modality. Ferati et al. (2018) [81] argue that if a particular modality or human sensory system works best in a given safety-critical situation, then the information should be provided in that modality. Bennett and Vijaygopal (2024) [64] used science-fiction prototyping as a technology foresight method, drawing on an imagination workshop with experts and a conjoint analysis with 661 people with ambulatory disabilities to assess their preferences for mobility and transportation technologies. The participants considered personal mobility assistive technology (either an automated wheelchair or an exoskeleton) to be the most important, followed by personal automation (autonomous vehicle or personal robot), and thirdly, by personal assistance technologies (real-time response versus augmented metaverse planning systems). AVs have been particularly appreciated by people with low transport self-efficacy, as they offer independence without the need for complex learning.

The literature highlights both the potential and the existing barriers to making AVs fully accessible for PLWD. Key challenges include the lack of standardized regulations, the need for early-stage accessibility considerations, and technological limitations in adapting to various disabilities. Additionally, the absence of accessibility regulations, the lack of involvement of PLWD in the design process, and technical challenges related to different types of disability (mobility, visual, auditory, cognitive) further hinder AV development tailored to their needs. An inclusive approach, with universal accessibility standards and more experimental research, is essential to ensure that AVs genuinely benefit PLWD [23]. Moving forward, integrating universal design principles, fostering user-centered research, and harmonizing accessibility regulations will be essential in ensuring that AVs can truly enhance mobility for all.

3.6.2. Developing Appropriate Internal Human–Machine Interfaces

A literature review on conventional vehicles has provided design guidelines for onboard interior design [85], aiming to create a repository applicable to the design of shared automated shuttles. In addition, participatory methods have been employed to generate interior HMI concepts. For instance, Brewer and Kameswaran (2018) [86] conducted design sessions with 15 visually impaired participants to develop an audio and tactile device to assist them in avoiding obstacles during control transition (i.e., takeover requests) in Level 3 or 4 automated vehicles. They also worked on general interaction with the AV (e.g., to give the destination), showing an interest in being able to interact naturally with the vehicle (i.e., talking to the vehicle naturally, as if it were a human).

The identification of design guidelines and concerns related to accessibility prompted the development and testing of specific internal and external HMIs (eHMIs) and features targeting PLWD. Mandujano Nava et al. (2017) [87] presented the process of creating a personal AV able to accommodate a wheelchair user and the resulting design concepts. However, no prototype or evaluation was conducted. In addition, the publication also reported interior HMIs designed to accommodate PLWD. Son et al. (2019) [88] designed a system to compensate for the inability of AV users with visual or auditory impairment to check their vehicle state in case of an accident. The system facilitated AV self-diagnosis and informed users through visual or auditory information, allowing interaction via speech or touch interface. While the system’s performance was validated through experiments, its usefulness and usability were not tested with PLWD. Castillo et al. (2014) [89] proposed another HMI specifically designed for PLWD, presenting the development process of a brain computer interface for PLWD to control AVs. This system, using electroencephalography, was laboratory tested with 19 participants, including 6 with physical impairment. Unfortunately, the authors do not specify whether this HMI was further tested or implemented.

Brinkley et al. (2018) [40] tested an interface designed to address the potential concerns identified in earlier research. Their experiment involved 20 participants with visual impairment in a Wizard-of-Oz AV [90] and allowed users to (1) specify the destination verbally or via a graphical interface; (2) provide auditory location cues during transit, designed to support situation awareness; (3) provide an oral description of the arrival location; and (4) monitor participants’ emotional state to adjust vehicle behavior. Most participants (93.75%) successfully completed the entire transit with positive perceptions of increased situation awareness, trust, perceived safety, and intention to use the system after the experience. Thus, this positive evaluation highlights the usefulness of studying PLWD’s views on AVs to enhance their accessibility. The integration of design guidelines and the testing of specific HMIs and features represent significant steps toward making AVs more inclusive for individuals with disabilities.

Expanding beyond mobility impairments, research has also explored the needs of older adults with mild to moderate cognitive impairment. These users demonstrated a preference for interfaces resembling familiar navigation systems (e.g., Google Maps), multimodal interactions (touch and voice commands), and personalized reminders [54]. Caregivers emphasized the need for remote monitoring, emotional support options (e.g., video calls), and emergency assistance tools. These findings suggest that user-centered interface design can mitigate cognitive load and improve AV adoption among older adults, although customization, trust-building measures, and caregiver integration remain critical for widespread acceptance.

A broader synthesis of research on AV HMI design was conducted by Brinkley et al. (2022) [91] to review state-of-the-art developments and their implications for AV accessibility. Their study emphasized the necessity of multimodal HMIs integrating voice interfaces, gesture interactions, and haptic feedback to accommodate users facing tactile or visual challenges. Moreover, HMIs should provide clear journey information and confirm accurate drop-off locations, particularly for users with navigation difficulties. These findings highlight the importance of tailoring AV interfaces to diverse needs to ensure that AV technology fosters an inclusive mobility experience.

3.6.3. External HMIs and Pedestrian Safety

The increasing automation of vehicles raises concerns about drivers potentially focusing less on the road because driving is delegated to the vehicle, or even if, with fully autonomous vehicles, there is no driver in the vehicle. However, the literature on pedestrians’ crossing behavior highlights the importance of implicit information, such as visual contact with the driver (e.g., [92,93,94,95]). To address the disappearance of this implicit communication, current research is focusing on the design of eHMIs (e.g., [96,97,98,99,100]). In response to this issue, Roundtree et al. (2020) [101] formulated design guidelines for inclusive eHMIs based on a literature review of pedestrian crossing behavior. These guidelines emphasize the need for implicit communication of AV intent, multimodal message delivery (visual and auditory), optimal display placement (three displays: front bumper and sides), and adaptive message formats (size, saliency, brightness, and audio cues based on ambient noise). The authors stressed the importance of validating these guidelines through real-world testing, an issue also raised by Colley et al. (2019) [102], who reviewed eHMI concepts from a universal design perspective. Their analysis found that 65% of studies relied solely on visual information, potentially excluding visually impaired pedestrians. This aligns with Dey et al. (2020) [103], who reported that only 29% of identified eHMI concepts incorporated multiple sensory modalities, with none adhering to universal design principles or testing with impaired participants. Such findings highlight a critical gap in considering PLWD during the early stages of AV interaction research (Tabattanon & D’Souza, 2019) [82]. Robert (2021) [104] also identified a gap in AV-pedestrian interactions, questioning how AV algorithms will accommodate impaired pedestrians, such as wheelchair users.

To further understand user preferences, Colley et al. (2020) [105] investigated eHMIs using a virtual reality setup with six visually impaired participants. They evaluated eight auditory HMI concepts and three tactile concepts via smartphones. The participants preferred explicit messages, spoken text, and standardized formats across manufacturers, but they expressed concerns about potential confusion between auditory cues and unrelated notifications (e.g., SMS alerts). Additional evaluations with 8 visually impaired and 25 sighted participants examined 5 street-crossing conditions, measuring cognitive load. Conditions featuring high-content vocal messages (e.g., “I’m stopping, you can cross”) resulted in the lowest cognitive load and were preferred over silent or ambiguous cues. These findings reinforce the importance of explicit, multimodal communication strategies for ensuring safe pedestrian–AV interactions.

4. Discussion

The advent and deployment of AVs represent an important shift in the field of urban and personal mobility and can present a huge opportunity for PLWD to improve their autonomy and social participation. However, to transform this opportunity into reality, several challenges concerning technological adaptation, legal and regulatory frameworks, and the implementation of appropriate accessibility standards remain. This discussion introduces the opportunities that AVs can represent for PLWD, alongside the critical barriers that must be addressed to allow the widespread adoption of AVs and make them a real opportunity for PLWD.

4.1. AV as an Opportunity for PLWD

The first objective of this literature review was to investigate why and how AVs could represent an opportunity for the mobility of PLWD. Firstly, the impact of AVs on the mobility of PLWD is contingent on the level of automation. Higher levels, specifically Level 4, hold greater promise for enhancing mobility, as lower levels still require individuals with driving abilities [35,59]. Furthermore, some AV advantages for the mobility of PLWD might vary when considering AVs used as private vehicles or as public transport. An AV as a private vehicle offers an additional mobility option, which is especially beneficial for PLWD who are unable to drive a conventional vehicle, and those who face barriers in accessing conventional public transport. This mode of transportation enhances mobility autonomy, eliminating the dependence on others for mobility needs [26,27,28,40]. Moreover, AVs in the professional context could offer new job opportunities for PLWD [50]. When employed as paratransit systems, AVs are seen as capable of providing better quality services with extended and flexible riding hours, and a broader service area [27,56]. Furthermore, the removal of human drivers could lead to a greater emphasis on passenger assistance and enhanced customer service [28]. Digital technology linked to AVs, such as onboard health status monitoring, can provide priority in traffic for accessing medical assistance [59]. In addition, the emergence of AVs has the potential to reduce transport costs for people with reduced mobility, particularly in shared mobility models. By optimizing vehicle allocation and reducing the dependence on specialized, costly transport services, AVs could offer an interesting alternative to current mobility options.

These application scenarios illustrate that AVs can address diverse mobility needs depending on context. While private AVs emphasize autonomy and independence, shared and public AV systems aim to reduce systemic barriers in current transit networks. The literature underscores that realizing these opportunities depends heavily on matching AV deployment strategies to the mobility profiles and financial means of different PLWD groups.

4.2. Challenges

The second objective was to highlight future challenges in ensuring AVs are an opportunity for PLWD. One key challenge revolves around the legal and regulatory framework for AV deployment, with concerns raised regarding licensing issues [26,29,49,50] and data privacy [27,50]. Considering these issues is essential to ensuring AV accessibility.

Another critical challenge is standardizing accessibility requirements for AVs [27] and collaboration between stakeholders [80]. Early consideration of PLWD needs in the design process is crucial to mitigate existing constraints [82]. In this context, the adoption of universal design principles is recommended [80,81]. Some of the identified challenges have already been partially addressed through targeted research. For instance, prototypes of wheelchair securement systems and adaptable cabin layouts have been developed [83], multimodal interfaces accessible to visually impaired users have been tested [63], and scenario-based policy planning has proposed regulatory reforms to improve access to AVs [27,50,59]. However, these solutions are still only being implemented in pilot studies and design proposals rather than being adopted more widely. Establishing accessibility standards has commenced through various methods, including gathering recommendations and guidelines from conventional transportation (e.g., [85]), consulting transportation specialists [56], and seeking input from PLWD (e.g., [68,69,70]). This collaborative approach serves to identify potential barriers and preferences, enabling anticipation and proactive measures to enhance accessibility to AVs. Insights garnered from PLWD opinions, complementing the existing literature on conventional vehicles, reveal shared concerns, such as equipment failure, with the broader population [40]. Additionally, specific concerns raised by PLWD aid in anticipating their future needs and the necessary constraints and features required [52,53]. These concerns and requests underscore the holistic approach required in addressing the entire transportation chain’s needs from an accessibility perspective before (i.e., booking app, payment system, pick-up area), during (i.e., give vehicle parking guidance), and after (i.e., drop-off area, orientation at the destination, verifying the arrival at the correct location) the trip. Moreover, PLWD also emphasize the necessity of considering the evolving role of the driver, such as potentially transforming towards customer service.

Cybersecurity and data protection are another major concern. AVs rely on the continuous exchange of data, and the collection of personal and medical information to improve accessibility features can expose people with disabilities to risks of unauthorized tracking, hacking, or the misuse of sensitive data. Strong regulatory frameworks are needed to ensure ethical data processing practices and avoid abuses.

Research has outlined the requirements to maximize transport accessibility for PLWD, based on a literature review on the barriers in transport encountered by them [106]. These requirements include accessible information, flexibility, safety, physically accessible design, reliability, economic predictability, reduced administration, and short, predictable travel times. Another issue regarding the introduction of AVs is to consider the specificities of PLWD in all their diversity, ranging from reduced mobility to visual and hearing impairments. Specific technical and non-technical assistance, such as wheelchair accommodation and a guide-dog-friendly design (e.g., to prevent guide dogs from sliding on floors), must be considered.

The literature has primarily focused on onboard interior accessibility, mainly considering this from the perspective of mobility-impaired users [85]. The literature has paid less attention to the accessibility of AVs from the pedestrian’s point of view. However, a small part of the literature considered the accessibility of AVs from the perspective of pedestrians via the design of eHMIs and allowed the formulation of design recommendations based on a literature review about pedestrian crossing behavior [101], interviews, and the testing of HMI concepts with virtual reality [105]. Nevertheless, these works remain marginal and need to be generalized.

A significant challenge to be addressed relates to the acceptance of AVs. PLWD’s perceptions of AVs influence their willingness to use them [75,76,77], and their concerns about AVs influence their acceptance [68]. Monitoring their acceptance is essential for adapting public communication about AVs and addressing evolving concerns. Overall, the literature reveals a growing but still fragmented effort to identify and solve these challenges. Emphasizing these early solutions, and the authors and institutions behind them, offers insight into the direction of future innovation and the collaborative pathways necessary to allow AV accessibility for PLWD.

4.3. Limitations and Future Research

It seems worth mentioning that the present literature review is limited as it was restricted to literature published in English and referenced in four databases. Additionally, this review primarily considered publications and conference proceedings, excluding possible contributions from private sector stakeholders actively working on AV accessibility solutions.

Despite these limitations, this literature review identifies avenues for future research. While some research has addressed the accessibility of eHMIs for PLWD with sensorial impairments, further efforts should focus on those with mental or cognitive impairments. Additionally, inquiries into the operational aspects of AV transport services, the role of onboard operators, and the potential for universal accessibility in public transport are crucial. Future studies should delve into solutions for the identified barriers in AV accessibility and address a broader spectrum of impairments, considering the entire diversity of PLWD. Furthermore, research into the economic viability of AV accessibility solutions, the effectiveness of pilot projects, and cross-country policy comparisons would provide valuable insights for widespread AV adoption.

5. Conclusions

This literature review examined how AVs could represent a promising route to improving mobility for people living with disabilities. It demonstrates that AVs, particularly those at automation Level 4 and above, could foster greater independence and reduce reliance on others, enabling wider participation in professional and social life. Whether used for personal travel or integrated into shared mobility networks, AVs can complement or even surpass current transport options, many of which have limited accessibility for this group. Additionally, this study emphasizes how AVs could reduce transportation costs, enhance service adaptability, and transform the role of paratransit by transitioning drivers to support roles and providing more personalized, user-friendly assistance.

This review also offers solid insight into the primary factors shaping AV adoption among PLWD. While there is clear optimism about AVs as tools for improved mobility, enduring concerns persist; chief among them are issues of safety, trustworthiness, lack of human support, and physical or digital accessibility. These concerns are diverse; they vary depending on disability type and individual user profiles. Attitudes toward AV use are influenced by demographic characteristics, prior transport experiences, and the level of confidence in AV technologies. The findings reinforce the need for universal design from the outset, inclusive design practices involving PLWD in development stages, and AV systems that are adaptive, accessible, and easy to use.

Finally, this review points to several areas requiring deeper investigation. Future research should broaden its reach to include disability types that are currently underrepresented, particularly cognitive disabilities, and conduct real-world AV trials involving diverse PLWD participants. Legal and regulatory structures also need to evolve, addressing licensing standards, data protection, and liability through an inclusive lens. In parallel, harmonized accessibility regulations and cost-benefit analyses of AV retrofitting or inclusive AV design should become focal points. Ultimately, a collaborative approach involving policymakers, industry leaders, and disability advocates will be critical to ensuring AVs evolve into an equitable, transformative mobility solution.