JRS
Zheng, J., Feng, Y., Tang, M., & Ramachandran, G. (2026). Perceptions of Smartphone Safety Alerts for Vulnerable Road Users at Urban Intersections. Journal of Road Safety, 37(3), 1–15. https://doi.org/10.33492/JRS-D-26-3-2881657

View more stats

Abstract

The safety of vulnerable road users (VRUs), including pedestrians and cyclists, remains a persistent challenge in complex urban traffic environments. While vehicle-to-everything (V2X) technologies have improved safety for vehicle occupants, VRUs are often not part of these systems and have not benefited from the technology-based safety gains. The widespread use of smartphones presents an accessible platform for delivering real-time safety alerts. This study investigated user perceptions, acceptance, and privacy attitudes toward smartphone-based vehicle-to-pedestrian (V2P) communication systems. A quantitative survey was conducted among road users (N = 153) with experience in Australia and China, capturing perspectives across different traffic environments and technology contexts. Semi-structured expert interviews were also conducted with participants holding academic or industry expertise (N = 5) to provide contextual insights into user expectations, interface design, and deployment considerations. Results show limited prior awareness of V2P applications but strong interest in adopting smartphone-based alerts for pedestrians, provided that systems are simple, responsive, and minimally distracting. High levels of acceptance were observed across participants, including those without prior experience of V2P alerts, suggesting that familiarity is not a prerequisite for adoption. The importance of trust and perceived control was highlighted as attitudes about privacy were distributed across unconditional acceptance, conditional acceptance, and refusal of location sharing. The study identifies key human-centred design priorities for smartphone-based safety systems and supports their feasibility when designed around usability, reliability, and privacy, addressing a gap in existing research that has predominantly focused on technical system performance. Findings provide practical insights for developing inclusive, scalable, and privacy-aware safety solutions for VRUs within intelligent transportation systems (ITS).

Key findings

  • Most participants were unaware of vehicle-to-pedestrian (V2P) safety applications before the survey.

  • Users prioritised ease of use and low distraction for safety alerts.

  • Privacy concerns significantly affected willingness to share location data.

  • Context-adaptive alerts (visual/audio/haptic) increased trust and usefulness.

Glossary

C-ITS   Cooperative Intelligent Transport Systems
C-V2X   Cellular Vehicle-to-Everything
DSRC   Dedicated Short-Range Communications
HCI   Human-Computer Interaction
ITS   Intelligent Transportation Systems
V2P   Vehicle-to-Pedestrian
V2X   Vehicle-to-Everything
VRU   Vulnerable Road User

Introduction

The rapid evolution of Intelligent Transport Systems (ITS) and vehicle-to-everything (V2X) communication has improved motor-vehicle safety for occupants by enabling real-time situational awareness and collision avoidance. This progress has not been equally achieved for people moving on the road network outside a motor vehicle. Pedestrians, cyclists and other micromobility users (e.g., e-scooters), collectively referred to as vulnerable road users or VRUs, continue to face disproportionately high risks in urban traffic environments (Olszewski et al., 2019). Current motor vehicle-centric approaches rely on dedicated short range communications (DSRC) or cellular protocols (A. R. Khan et al., 2021), but VRUs generally lack access to such systems as a result, the V2X safety gains, particularly in dense traffic environments where limited visibility and delayed reaction times are critical factors have not be realised for VRUs (Kutela et al., 2022; Malone et al., 2017). Evidence from Queensland crash data further shows that a substantial proportion of traffic incidents occur at intersections, including cross intersections, T-junctions, and roundabouts, reinforcing the importance of intersection-focused safety considerations in urban environments (Queensland Department of Transport and Main Roads (TMR), 2024).

The widespread use of smartphones provides a low-cost and scalable pathway for extending V2X capabilities to VRUs (Kegenbekov & Saparova, 2022; Mishra & Kertesz, 2020). In particular, smartphone-based V2P alerts can complement existing safety measures by providing timely situational awareness to road users who otherwise lack access to motor vehicle-based sensing and communication systems. Crash analysis research has highlighted that many pedestrian and cyclist crashes occur at urban intersections, where complex interactions, limited visibility, and short reaction times increase risk (Oregon Department of Transportation, Research Section (ODOT), 2017).

The effectiveness of such systems depends on both technical feasibility and human-centred factors. Frequent or irrelevant warnings may lead to alarm fatigue (Alyamani et al., 2024; Sutton & Wood, 2025) and cognitive overload (I. Khan & Khusro, 2020; Wu & Chen, 2023), while privacy concerns related to continuous location sharing may reduce user trust and willingness to adopt such systems (Baron & Musolesi, 2020). In addition, poorly designed alert interfaces can increase distraction or delay responses, potentially undermining safety benefits (Horberry et al., 2021).

Despite ongoing advances in road-safety technologies, large-scale initiatives continue to focus on motor vehicles, leaving VRU needs under-represented in the broader ITS agenda (Chen et al., 2017; World Health Organization, 2023). As a result, many existing interventions prioritise motor vehicle-based or motor vehicle focused infrastructure-led solutions, which may not adequately address the short-range, dynamic interaction risks faced by VRUs in complex urban environments, particularly at intersections.

This study addressed these gaps by examining how smartphone-based V2P systems can be designed and accepted to support VRU safety. The study was guided by three research questions:

  • Are smartphone-based alerts acceptable in everyday use?

  • How do users differ in their preferences for data sharing and alert design?

  • How can interface design support attention and trust without increasing cognitive load?

Method

This study used an exploratory, data-oriented approach that combined quantitative survey data and qualitative interview insights to examine user acceptance, usability, and privacy attitudes toward smartphone-based safety alerts for VRUs. The research consisted of two primary components: an online survey to capture quantitative patterns of user preferences and perceptions, and semi-structured interviews to provide qualitative insights into the interpretation of these patterns.

The survey was the primary data source to identify general patterns and trends in user responses while the interviews are used to interpret these findings, explore underlying reasons, and provide domain-specific context for observed behaviours. This approach ensures that qualitative insights complement and contextualise the quantitative results, rather than serving as direct validation.

Survey procedure

Survey design rationale

The design of the survey instrument was informed by recent literature on user acceptance, alert modality design, communication reliability, and privacy in smartphone based ITS. Frequent or irrelevant warnings may lead to alert fatigue and reduced trust in safety systems (Alyamani et al., 2024), while privacy concerns in location-based systems may reduce willingness to adopt such technologies (Poikela, 2020).

Further, the structure and operationalisation of key variables in the survey were informed by synthesised findings from previous works on V2X and V2P communication architectures (Zoghlami et al., 2023), smartphone-based safety applications (Kegenbekov & Saparova, 2022; Mishra & Kertesz, 2020), and privacy-preserving frameworks such as pseudonym-based authentication and the PRESERVE architecture (Neven et al., 2017; Santa et al., 2014), and user adoption of mobile warning systems (Fischer-Preßler et al., 2022).

The survey incorporated structured items assessing perceived usefulness, reliability, privacy sensitivity, and alert fatigue tolerance. Respondents were asked to indicate their preferences regarding alert frequency, data-sharing conditions, and the balance between safety benefits and perceived intrusiveness. This design ensured that the survey captured both functional expectations and psychological thresholds shaping user acceptance of smartphone-based V2P safety applications.

The survey instrument was designed to capture both behavioural and perceptual dimensions of user interaction with smartphone-based V2P safety systems with a focus on three central domains aligned with the research questions: 1) user experience and system feasibility; 2) communication and alert preferences, and; 3) privacy and data-sharing attitudes. Questions on system feasibility assessed prior exposure to safety applications, perceived usefulness, and device compatibility. Questions on alert design explored preferred modalities, alert frequency tolerance, and relevance sensitivity to identify thresholds for trust and potential alert fatigue. Questions on privacy evaluated users’ comfort with data sharing, control preferences, and conditional consent under different disclosure scenarios.

Each domain consisted of targeted items combining single-choice and multiple-choice questions. Question phrasing prioritised clarity, neutrality, and minimal cognitive load. Scenarios were written in plain language to ensure accessibility for non-specialist participants, avoiding technical descriptions of V2X protocols or system architectures. The identical questionnaire was administered in both Australia and China without region-specific adaptations to ensure consistency across contexts. Participants in Australia completed the English-language questionnaire while participants in China completed a bilingual (Chinese-English) version containing equivalent questionnaire content. The questionnaire followed a progressive structure, starting from general mobility habits to specific attitudes toward smartphone-based safety alerts, to reduce priming effects and improve response validity.

Pilot testing with a small group of volunteer respondents (n = 20) was conducted to refine question wording, option balance, and survey length, and ensure completion within 10-15 minutes. This iterative process improved reliability, content validity, and consistency across participant groups. The full questionnaire is included in Appendix A.

Survey administration

Structured online surveys were administered via the Qualtrics platform using a convenience sampling approach to drivers, cyclists, and pedestrians with road-use experience in Australia and China. Participants were required to be at least 18 years old and to have prior experience using smartphone-based navigation or mobility-related applications. The survey was primarily distributed among university student populations, supplemented by a smaller number of participants from the general working population. Distribution was conducted through university mailing lists, social media platforms, and personal networks.

To monitor the reliability and stability of the survey instrument during data collection, an incremental sampling approach was adopted. Interim analyses were conducted after approximately 50 and 100 responses to assess whether emerging patterns remained consistent across demographic groups and key survey constructs. The distribution of responses, including preferred alert modalities, privacy attitudes, and perceived road-risk scenarios, showed no substantial deviations across these stages. This consistency suggests that the questionnaire functioned as intended and that response patterns remained stable throughout the data collection process.

The questionnaire included items measuring respondents’ willingness to adopt smartphone-based safety alerts, perceived usefulness, tolerance for alert frequency and latency, and privacy preferences regarding location sharing. Additional items explored attitudes toward different alert modalities (auditory, visual, and haptic) as well as perceptions of trust and potential distraction.

Interview procedure

To complement the survey results, semi-structured expert interviews were conducted with participants who have academic or industry expertise in road safety and intelligent transport systems. A total of five participants (N = 5) were included, recruited through academic and professional networks, including university contacts and industry collaborations, with backgrounds spanning relevant disciplines and practical experience.

The interviews were designed to examine observed trends in survey responses, explore underlying reasons for user responses, and provide domain-specific insights into factors influencing user preferences, rather than to directly validate user responses. Each interview lasted approximately 25-40 minutes and was conducted either face-to-face or online. All interviews were audio-recorded and transcribed verbatim by the first author to ensure accuracy and familiarity with the data.

Data analysis

Survey responses were analysed to identify patterns in user preferences and acceptance behaviours. Descriptive statistics (percentages and frequency distributions) were used to summarise responses on alert design, data privacy, and usability.

Chi-square tests were used to examine differences between respondent groups across key variables, including alert preferences and privacy attitudes. An additional odds ratio (OR) analysis was conducted to explore the association between prior experience and willingness to adopt V2P alerts. Given the exploratory nature of the study and the structure of the dataset, more complex multivariable models were not applied. Analyses were based on valid responses for each item, with variations in sample size reflecting response availability, role-specific questions, and partial completion. Effect sizes (Cramér’s V) were reported to support interpretation.

Interview data were analysed using thematic analysis. Transcripts were reviewed to identify recurring themes, which were systematically compared with the survey findings to provide contextual interpretation. The qualitative insights were used to clarify ambiguous responses and explain differences across user groups, rather than to validate survey results.

Results

In total, 153 respondents completed the surveys, including 101 participants from Australia (66%) and 52 participants from China (34%). The sample included drivers, cyclists, and pedestrians from both countries. Participant characteristics for the Australian and Chinese cohorts are summarised in Table 1.

Survey

Table 1 summarises the key quantitative findings from the online survey. To present the results clearly and systematically, the table is organised into four major sections reflecting the structure of the questionnaire and the different dimensions of user responses. The first section presents participant demographics, including age distribution, countries where respondents gained their road-use experience, and their reported road-user roles. The second section reports responses related to VRU safety awareness, including willingness to engage with safety applications, identification of collision-prone environments, and acceptable hardware costs. The third section summarises user acceptance and preferences, including preferred communication methods for V2P interaction, preferred alert formats, openness to receiving mobile alerts, prior experience with safety applications, and attitudes toward privacy and location sharing. The fourth section presents responses related to V2P interaction and alert usability, including undesirable warning situations, preferred alert modalities across regions, tolerance for false alarms, and differences in information needs across road-user roles. The results reported in Table 1 report the distribution of responses across these categories. To supplement the descriptive results, chi-square tests were conducted to examine whether key preferences differed across regions. No statistically significant differences were observed between respondents in Australia and China in:

  • Alert system preferences (app-based, navigation-integrated, hybrid, no preference, and other) (χ²(4, N = 93) = 7.16, p = 0.128, Cramér’s V = 0.28)

  • Privacy attitudes (unconditional, conditional, refusal, none) also did not significantly differ between regions (χ²(3, N = 99) = 3.30, p = 0.347, Cramér’s V = 0.18)

For multiple-response items such as collision-prone environments, responses were analysed descriptively using frequency distributions. Intersections were among the most frequently identified high-risk scenarios, reported by a similar proportion of respondents in Australia (46%) and in China (50%).

An additional analysis based on a 2×2 contingency table explored the association between prior experience and willingness to adopt safety alerts. High levels of willingness, indicating acceptance, were observed in both experienced users (85%) and inexperienced users (79%). The corresponding odds ratio (OR = 1.56) indicates a slight difference between groups, suggesting that prior experience may have a limited influence on willingness.

Sample sizes vary across analyses due to role-specific questions, multiple-response items, and partial survey responses, as described in the methodology.

Table 1.Survey results and findings
Section Variable Responses Australia (n = 101) China (n = 52) All (N=153) Findings
Participants Age (years) <18 5% Respondents were predominantly young adults with road-use experience mainly in Australia, often across multiple roles (primarily pedestrians and drivers), with fewer cyclists.
18-40 84%
40-60 9%
60+ 2%
Road user experience in Australia or other countries Australia 45%
Other countries 38%
Both 17%
Road user category (multiple responses) Pedestrian 65%
Driver 65%
Cyclist 29%
VRU safety awareness Top 3 Dangers / high risk conditions (multiple responses) Poor visibility 52% - Intersections and visibility-related conditions were identified as high-risk scenarios, alongside driver distraction and traffic rule non-compliance as additional concerns.
Not following traffic rules - 53%
Driver distraction 49% 50%
Intersection 46% 50%
Willingness to test the safety app and device No 63% Participants showed interest in safety technologies but preferred stable, fully developed systems over prototypes.
Yes 37%
Acceptable additional hardware costs (AUD) Less than $300 84% Most respondents were cost-sensitive, preferring any additional hardware to remain within a low-cost range.
$300-$500 14%
$500-$1,000 2%
User acceptance and preferences Preferred V2P interaction methods for enhancing safety Short-range 35% Preferences were diverse, with no single method fitting all connectivity contexts.
Combination 35%
Internet-based 20%
No interconnection 6%
Other 4%
Preferred safety warning types to receive (multiple responses) Integration 55% 46% Navigation-integrated warnings were most preferred, indicating expectations for seamless, low-effort use.
Hybrid 37% 27%
App 5% 12%
Acceptable safety alerts or warning messages from a mobile device Willing 74% Most participants were willing to receive mobile safety alerts.
Neutral 21%
Unwilling 5%
Experience of using safety mobile apps No 86% 59% Prior experience with safety apps was limited despite general willingness to use them.
Yes 14% 41%
Road users' concerns about location privacy Unconditional consent 29% Privacy attitudes were relatively distributed, indicating no dominant preference and the need for transparent data-handling practices.
Conditional consent 37%
Refusal 33%
None of above 1%
V2P interaction Top 3 User's undesirable safety warning circumstances (multiple responses) Too frequent 71% Overly frequent, unimportant, and insufficiently noticeable alerts were the main factors reducing effectiveness.
Not important 52%
Not noticeable 44%
Top 3 Preferable safety alert form (multiple responses) Visual displays 54% - Sound was the most consistently preferred alert modality across both regions, while visual cues (e.g., on-screen prompts or LED indicators) were valued as supplementary support.
Sound 49% 52%
LED light - 35%
Pedestrian crossing signals 38% 32%
Vibration - 32%
Top 3 Tolerance and acceptance of false alarms triggered Tolerate occasional 26% 19% Tolerance for false alarms was generally low, with repeated inaccuracies leading to reduced trust or discontinuation of use.
Reduce trust 20% 35%
Turn off 18% 19%
Top 3 Preferred message when crossing roads and intersections (multiple responses) Driver
  • Sudden appearance of VRUs
 76% Preferred alert scenarios varied across road-user groups, with drivers emphasising sudden VRU appearance, adverse weather, and unsafe behaviours by other vehicles.
  • Adverse weather
 76%
  • Dangerous driving behaviours by other vehicles
 62%
Cyclist
  • Turning vehicles
 64%
  • Accident-prone area alerts
 54%
  • Vehicles crossing the road
 31%
Pedestrian
  • Traffic signal status
 65%
  • Vehicle speed
 63%
  • Vehicle distance
 63%

Interview findings

Semi-structured expert interviews were conducted to provide additional insights and to complement the survey findings by exploring underlying reasoning, contextual factors, and expert perspectives on smartphone-based safety alerts. Qualitative findings were analysed thematically, identifying recurring patterns across participants. Table 2 presents a summary of these findings and shows interviewees identified intersections, blind spots, slopes, and low-visibility conditions as key scenarios where alerts would be most beneficial. These observations were broadly consistent with patterns observed in the survey results, in which intersections and visibility limitations were also frequently reported.

Participants noted that expected warning lead-times differ by road-user type, with drivers and cyclists requiring longer advance notice for braking or manoeuvring, and pedestrians preferring more immediate cues. Reliability was consistently highlighted, with delayed or irrelevant alerts identified as factors that could reduce user trust and system usage.

Regarding alert delivery, participants generally preferred multimodal alerts, particularly combinations of sound and visual cues, while vibration was considered useful in specific contexts. Several interviewees also indicated that alert frequency and alert type should be adjustable, allowing users to tailor the system to their preferences. These views aligned with trends observed in the survey responses regarding alert modality and user expectations.

Attitudes about privacy varied. Some participants expressed minimal concern about real-time location sharing, while others preferred stricter data minimisation, context-dependent activation, or avoidance of persistent identity linkage. These perspectives provide additional context for the diversity of privacy preferences responses in the survey data.

Views on activation modes were mixed: some participants preferred manual opt-in control, while others favoured automatic activation for convenience. All interviewees supported the inclusion of an in-app feedback mechanism to report alert usefulness and support system refinement.

Table 2.Interview results and findings
Theme Summary of key findings
Perceived necessity and use scenarios
  • Alerts were considered most useful in intersections, blind spots, slopes, and low-visibility or adverse weather conditions.
  • Alerts were viewed as complementary to visual checks and vehicle sensors.
Timing expectations
  • Warning lead time should vary by road-user type.
  • Longer lead times were preferred for drivers and cyclists due to braking and manoeuvring needs, while pedestrians preferred shorter, immediate alerts.
  • Although specific expectations varied, the overall pattern suggested alert timing should align with users’ movement speed and reaction needs.
  • Delayed alerts were associated with reduced trust.
Reliability concerns
  • False alarms and repeated irrelevant warnings were identified as major sources of distrust, often leading to reduced engagement or discontinuation.
Preferred alert types
  • Multimodal alerts (sound + visual) were generally preferred.
  • Vibration was useful in specific contexts.
  • Preferences varied between customisable and simple fixed alert styles.
Experience with existing tools
  • Limited experience with dedicated safety applications.
  • Motor vehicle-based warnings were perceived as useful but sometimes inaccurate.
  • Navigation applications were familiar but not always reliable.
Data privacy
  • Privacy attitudes ranged from unconcerned to cautious.
  • Participants preferred limited data collection, avoidance of long-term identity tracking, and greater transparency and control.
User control and activation
  • Preferences included manual activation and explicit opt-in mechanisms, although some participants favoured automatic activation for convenience.
Feedback mechanisms
  • All participants supported the inclusion of in-app feedback functions to report alert usefulness and improve system performance.

Discussion

Integration with existing literature

The findings extend prior smartphone-based V2P research by providing empirical evidence on user perceptions, trust, and interaction behaviours in real-world contexts. While earlier research largely focused on technical feasibility and system architectures (Gelbal et al., 2024; Zoghlami et al., 2023) and privacy architectures (Kumar et al., 2021; Neven et al., 2017), this study adds a human-centred perspective by incorporating behavioural insights and contextual variability into design considerations.

Recent studies have examined V2P communication architectures and system-level implementations, including smartphone-based and cellular technology that uses mobile communication networks (i.e., 4G and 5G networks for C-V2X-enabled warning systems (Gelbal et al., 2024; Malik et al., 2020; Zhang et al., 2023). Advances in 5G-enabled V2X communication further support this direction, with improvements in latency and system responsiveness reported in VRU safety applications (Zoghlami et al., 2023). Despite these developments, there is a continuing research focus on communication performance and system architecture, with relatively limited attention to user experience, acceptance, and behavioural responses in real-world contexts. This study addresses that gap by examining user perceptions, trust, and interaction behaviours, linking technical feasibility with practical deployment considerations.

User perceptions of risk and system necessity

Survey results identified intersections, poor visibility, and driver distraction as the primary concerns for VRUs. The prominence of intersections highlights their importance as key deployment contexts for V2P systems, where dynamic movements and crossing conflicts are frequent. Interview participants further emphasised conditions such as blind spots, steep slopes, and adverse weather, reinforcing the situational relevance of smartphone-based alerts for short, high-risk exposure periods.

These findings are consistent with prior work suggesting that smartphones can serve as accessible platforms for real-time safety alerts (Kegenbekov & Saparova, 2022) and with collision-avoidance frameworks identifying intersections and low-visibility zones as latency-critical contexts for VRU protection (Hasan et al., 2018). Survey results also showed that perceptions of intersection-related risk did not differ significantly across pedestrians, cyclists, and drivers, supporting the use of intersection-focused V2P systems as a common design context.

Additional analysis indicates that willingness to adopt smartphone-based safety alerts remains high regardless of prior experience, suggesting that prior familiarity is not a prerequisite for adoption.

Alert modalities and Human-Computer Interaction (HCI)

Both survey and interview data revealed a strong preference for combined sound and visual alerts, suggesting that multimodal cues can improve user awareness and response in safety-critical situations. This is consistent with Baldwin et al. (2012), who highlight that the effectiveness of multimodal warnings in complex environments depends on task demands, information characteristics, and system design. Participants valued customisable alert modes but cautioned against overly adaptive or intrusive notifications, identifying concerns consistent with recent findings on alert fatigue and reduced trust due to irrelevant or excessive warnings (Alyamani et al., 2024). This suggests that multimodality should be combined with simplicity and user control. Smartphone-based systems can therefore leverage multimodal feedback and context-aware adaptation to deliver timely and relevant notifications without overwhelming the user (Mishra & Kertesz, 2020). Similarly, prior HCI research reported that multimodal cues can support situational awareness while managing cognitive load (I. Khan & Khusro, 2020; Sodnik et al., 2008; Wu & Chen, 2023). Studies of mobile Cooperative Intelligent Transport Systems (C-ITS) applications, which support communication between vehicles, infrastructure, and road users, further indicate that user experience, perceived usefulness, and interface simplicity influence both acceptance and continued use of safety systems (Burger & Guna, 2024), while perceived reliability remains critical for sustained engagement (Voinea et al., 2020). Variations in modality preferences across different contexts also support the need for adaptive HCI design that responds to user behaviour rather than relying on fixed interaction models (Höök, 2000).

System format and integration preferences

The majority preference for navigation-integrated or hybrid applications suggests that users favour safety features embedded within tools they already use. This is consistent with the Technology Acceptance Model (Davis, 1989) that highlights the role of perceived ease of use and compatibility in adoption. As one interviewee (male, 25-35 years, with experience walking and driving in Australia) noted, “It should work like Google Maps, not another app I need to open”. Such expectations align with existing V2X implementations, such as Smart Walk Assistant (Khosravi et al., 2018) that integrates pedestrian-vehicle interactions into existing mobility systems. These findings point to the importance of minimal setup, automatic activation, and intuitive integration within navigation platforms to support real-world adoption. The absence of statistically significant regional differences in alert-format preferences suggests that users in both Australia and China share similar expectations regarding integration with existing navigation tools, indicating that low-effort integration may be a broadly applicable design priority.

Reliability, frequency and trust dynamics

Alert reliability emerged as a key determinant of continued use. Quantitative findings show that most respondents (71%) ignored alerts when too frequent, while half (52%) disengaged when alerts were perceived as unimportant. This is consistent with prior evidence that excessive and irrelevant notifications can erode user trust (Alyamani et al., 2024), highlighting the need for adaptive frequency control mechanisms to mitigate alert fatigue.

Interview participants further elaborated that even a small number of false alarms could reduce trust in the system (a view expressed by three male interviewees aged 25-35 years with pedestrian and cycling experience in Australia) emphasising the importance of precise, context-aware warning logic. Future implementations may benefit from incorporating frequency moderation and graded urgency cues to balance timeliness and cognitive comfort.

Privacy and data-sharing

Attitudes about privacy were distributed across unconditional acceptance, conditional consent, and outright refusal, suggesting that privacy preferences remain highly individualised and context-dependent. Interview participants responses indicated a limited understanding of underlying data-handling mechanisms, suggesting that technical safeguards alone may not translate into perceived trustworthiness.

The pattern of conditional consent aligns with privacy calculus theory (Dinev & Hart, 2006), where users weigh perceived risks and benefits when deciding whether to share data. This is also consistent with prior research on mobile privacy decision-making, which highlights the importance of transparency and user-understandable controls in shaping trust and acceptance (Betzing et al., 2020). Together, these findings suggest that transparency and perceived control may be more influential than encryption type or protocol complexity. Behavioural models such as Protection Motivation Theory similarly indicate that adoption decisions are shaped by perceived risks and benefits (Fischer-Preßler et al., 2022).

Differences in responses between Australian and Chinese respondents were not statistically significant for private attitude distributions nor alert system preferences although findings should be interpreted with caution, given the exploratory nature of the study. Further, user acceptance appears to be influenced by perceived control and transparency, in addition to technical considerations.

Design and policy implications

Findings suggest that alert modality, alert relevance and frequency, and privacy control mechanisms shape user trust, which in turn influences the adoption of smartphone-based V2P systems at urban intersections. The empirical evidence shows that multimodal alerts drive perceived usefulness, false-alarm tolerance shapes reliability expectations, and transparent data governance supports privacy acceptance. Together, these factors shape users’ perceptions of system credibility, intrusiveness, and suitability for everyday mobility, which in turn influence adoption decisions.

The findings provide actionable insights for smartphone-based V2P systems:

  • Multimodal and minimal interfaces can improve hazard detection without increasing distraction.

  • Adaptive frequency control can reduce alert fatigue while maintaining perceived reliability.

  • Transparent privacy management and clear opt-in mechanisms can support long-term acceptance.

  • Integration with existing navigation platforms can reduce usability barriers.

These implications point to practical design priorities for scalable and user-acceptable safety systems. Smartphone-based V2P systems can serve as a complementary approach to existing motor vehicle- and infrastructure-based safety measures, particularly in addressing conflicts at urban intersections.

Study strengths and limitations

A key strength of this study lies in its combination of survey data and expert interviews to capture both quantitative trends and contextual insights, providing a more comprehensive understanding of user perceptions, preferences, and privacy concerns. However, the relatively small and demographically skewed sample limits the generalisability of the findings, particularly for older adults, children, and users with accessibility needs. The concentration of participants aged 18-40 years further constrains the applicability of the results, and the findings should therefore be interpreted as indicative rather than representative. The limited number of interview participants (N = 5), primarily representing expert perspectives, also restricts the diversity of qualitative insights.

Future work should incorporate larger and more diverse samples, as well as field trials and prototype-based evaluations to validate user responses and evaluate system performance under realistic road conditions.

Conclusions

This study examined user perceptions of smartphone-based V2P warning systems through survey data and expert interviews (N = 5) conducted in Australia and China. The findings indicate strong support for using smartphones as accessible safety tools, provided that alerts are reliable, context-aware, and minimally intrusive. Participants preferred multimodal notifications combining sound and visual cues, integration with existing navigation tools, and transparent data-sharing options. Privacy attitudes were distributed across unconditional acceptance, conditional acceptance, and refusal, highlighting the role of trust and perceived control in shaping user acceptance.

The results also indicate that willingness to adopt smartphone-based safety alerts remains high regardless of prior experience, suggesting that familiarity is not a prerequisite for adoption. This points to relatively low barriers to uptake, with perceived usefulness, reliability, and system design playing a more important role than prior exposure to similar technologies.

The findings support the feasibility of smartphone-based V2P communication when systems are designed with attention to usability, privacy, and local conditions. User acceptance appears to be influenced by multiple factors, including system reliability, interface design, and perceived relevance in real-world scenarios.

Future work should extend to larger and more diverse samples, incorporate field trials to observe real-world behaviour, and explore adaptive alert strategies to reduce alert fatigue. Further research on privacy-preserving mechanisms, such as lightweight anonymisation and transparent consent models, will also be important for maintaining public trust. These directions support the development of inclusive, low-cost, and scalable smartphone-based safety solutions for VRUs within ITS.

AI tools

ChatGPT (OpenAI) was used solely to assist with English expression and clarity during manuscript preparation. It was not used for data collection, data analysis, interpretation of results, or generation of research findings or conclusions. The authors reviewed and approved all content in the final manuscript.

Acknowledgements

The authors thanks Dr. Jinglan Zhang for her helpful suggestions related to the survey process.

Author contributions

Jingfei Zheng: Conceptualisation, methodology, data collection, formal analysis, investigation, visualisation, writing original draft. Prof. Yanming Feng: Supervision, project guidance, methodology review, review & editing. Dr. Maolin Tang & Dr. Gowri Ramachandran: Review & editing, technical guidance, supervision support.

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Human Research Ethics Review

All procedures involving human participants were approved by the Queensland University of Technology (QUT) Human Research Ethics Committee (HREC) and conducted in accordance with the Declaration of Helsinki. Ethical approval was granted for both the survey and interview components of this study under QUT HREC approval numbers 2000000396 (online survey) and 3210 (semi-structured interviews).

All participants were provided with an information sheet outlining the study purpose, procedures, and confidentiality measures before giving informed consent. Participation was voluntary, and participants could withdraw at any time without penalty. Survey responses were anonymised, and no identifying information was collected. Interview participants provided written or electronic consent before participation. All data were securely stored and used solely for academic research purposes in accordance with QUT’s Information Privacy Policy.

Data availability statement

All materials, survey instruments, anonymised data, and analysis protocols used in this study are available from the corresponding author upon reasonable request.

Conflicts of interest

The authors declare that there are no conflicts of interest.

Submitted: December 12, 2025 AEST

Accepted: May 12, 2026 AEST

References

Alyamani, H., Yang, Y., Noyce, D., Chitturi, M., & Fawaz, K. (2024). Evaluating the Impact of Warning Modalities and False Alarms in Pedestrian Crossing Alert System. arXiv Preprint arXiv:2410.06388. https:/​/​doi.org/​10.48550/​arXiv.2410.06388
Google Scholar
Baldwin, C. L., Spence, C., Bliss, J. P., Brill, J. C., Wogalter, M. S., Mayhorn, C. B., & Ferris, T. K. (2012). Multimodal cueing: The relative benefits of the auditory, visual, and tactile channels in complex environments. In Proceedings of the Human Factors and Ergonomics Society Annual Meeting (Vol. 56, Issue 1, pp. 1431–1435). SAGE Publications. https:/​/​doi.org/​10.1177/​1071181312561404
Google Scholar
Baron, B., & Musolesi, M. (2020). Where you go matters: a study on the privacy implications of continuous location tracking. Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies, 4(4), 1–32. https:/​/​doi.org/​10.1145/​3432699
Google Scholar
Betzing, J. H., Tietz, M., vom Brocke, J., & Becker, J. (2020). The impact of transparency on mobile privacy decision making. Electronic Markets, 30(3), 607–625. https:/​/​doi.org/​10.1007/​s12525-019-00332-3
Google Scholar
Burger, G., & Guna, J. (2024). Enhancing driving safety through user experience evaluation of the C-ITS mobile application: a case study of the DARS traffic plus app in a driving simulator environment. Sensors, 24(15), 4948. https:/​/​doi.org/​10.3390/​s24154948
Google Scholar
Chen, B. H., Huang, S. C., & Tsai, W. H. (2017). Eliminating driving distractions: Human-computer interaction with built-in applications. IEEE Vehicular Technology Magazine, 12(1), 20–29. https:/​/​doi.org/​10.1109/​MVT.2016.2625331
Google Scholar
Davis, F. D. (1989). Perceived usefulness, perceived ease of use, and user acceptance of information technology. MIS Quarterly, 13(3), 319–340. https:/​/​doi.org/​10.2307/​249008
Google Scholar
Dinev, T., & Hart, P. (2006). An extended privacy calculus model for e-commerce transactions. Information Systems Research, 17(1), 61–80. https:/​/​doi.org/​10.1287/​isre.1060.0080
Google Scholar
Fischer-Preßler, D., Bonaretti, D., & Fischbach, K. (2022). A protection-motivation perspective to explain intention to use and continue to use mobile warning systems. Business & Information Systems Engineering, 64(2), 167–182. https:/​/​doi.org/​10.1007/​s12599-021-00704-0
Google Scholar
Gelbal, S. Y., Aksun-Guvenc, B., & Guvenc, L. (2024). Vulnerable road user safety using mobile phones with vehicle-to-vru communication. Electronics, 13(2), 331. https:/​/​doi.org/​10.3390/​electronics13020331
Google Scholar
Hasan, K. F., Wang, C., Feng, Y., & Tian, Y. C. (2018). Time synchronization in vehicular ad-hoc networks: A survey on theory and practice. Vehicular Communications, 14, 39–51. https:/​/​doi.org/​10.1016/​j.vehcom.2018.09.001
Google Scholar
Höök, K. (2000). Steps to take before intelligent user interfaces become real. Interacting with Computers, 12(4), 409–426. https:/​/​doi.org/​10.1016/​S0953-5438(99)00006-5
Google Scholar
Horberry, T., Mulvihill, C., Fitzharris, M., Lawrence, B., Lenné, M., Kuo, J., & Wood, D. (2021). Human-centered design for an in-vehicle truck driver fatigue and distraction warning system. IEEE Transactions on Intelligent Transportation Systems, 23(6), 5350–5359. https:/​/​doi.org/​10.1109/​TITS.2021.3053096
Google Scholar
Kegenbekov, Z., & Saparova, A. (2022). Using the MQTT protocol to transmit vehicle telemetry data. Transportation Research Procedia, 61, 410–417. https:/​/​doi.org/​10.1016/​j.trpro.2022.01.067
Google Scholar
Khan, A. R., Jamlos, M. F., Osman, N., Ishak, M. I., Dzaharudin, F., Yeow, Y. K., & Khairi, K. A. (2021). DSRC technology in vehicle-to-vehicle (V2V) and vehicle-to-infrastructure (V2I) IoT system for intelligent transportation system (ITS): A review. In Recent trends in mechatronics towards industry 4.0: selected articles from iM3F 2020, Malaysia (pp. 97–106). https:/​/​doi.org/​10.1007/​978-981-33-4597-3_10
Google Scholar
Khan, I., & Khusro, S. (2020). Towards the design of context-aware adaptive user interfaces to minimize drivers’ distractions. Mobile Information Systems, 2020(1), 8858886. https:/​/​doi.org/​10.1155/​2020/​8858886
Google Scholar
Khosravi, S., Beak, B., Head, K. L., & Saleem, F. (2018). Assistive system to improve pedestrians’ safety and mobility in a connected vehicle technology environment. Transportation Research Record, 2672(19), 145–156. https:/​/​doi.org/​10.1177/​0361198118783598
Google Scholar
Kumar, R., Kumar, P., Tripathi, R., Gupta, G. P., Kumar, N., & Hassan, M. M. (2021). A privacy-preserving-based secure framework using blockchain-enabled deep-learning in cooperative intelligent transport system. IEEE Transactions on Intelligent Transportation Systems, 23(9), 16492–16503. https:/​/​doi.org/​10.1109/​TITS.2021.3098636
Google Scholar
Kutela, B., Das, S., & Dadashova, B. (2022). Mining patterns of autonomous vehicle crashes involving vulnerable road users to understand the associated factors. Accident Analysis & Prevention, 165, 106473. https:/​/​doi.org/​10.1016/​j.aap.2021.106473
Google Scholar
Malik, R. Q., Ramli, K. N., Kareem, Z. H., Habelalmatee, M. I., Abbas, A. H., & Alamoody, A. (2020). An overview on V2P communication system: Architecture and application. In 2020 3rd International Conference on Engineering Technology and its Applications (IICETA) (pp. 174–178). IEEE. https:/​/​doi.org/​10.1109/​IICETA50496.2020.9318863
Google Scholar
Malone, K., Silla, A., Johanssen, C., & Bell, D. (2017). Safety, mobility and comfort assessment methodologies of intelligent transport systems for vulnerable road users. European Transport Research Review, 9(2), 21. https:/​/​doi.org/​10.1007/​s12544-017-0235-y
Google Scholar
Mishra, B., & Kertesz, A. (2020). The use of MQTT in M2M and IoT systems: A survey. IEEE Access, 8, 201071–201086. https:/​/​doi.org/​10.1109/​ACCESS.2020.3035849
Google Scholar
Neven, G., Baldini, G., Camenisch, J., & Neisse, R. (2017). Privacy-preserving attribute-based credentials in cooperative intelligent transport systems. In 2017 IEEE Vehicular Networking Conference (VNC) (pp. 131–138). IEEE. https:/​/​doi.org/​10.1109/​VNC.2017.8275631
Google Scholar
Olszewski, P., Szagała, P., Rabczenko, D., & Zielińska, A. (2019). Investigating safety of vulnerable road users in selected EU countries. Journal of Safety Research, 68, 49–57. https:/​/​doi.org/​10.1016/​j.jsr.2018.12.001
Google Scholar
Oregon Department of Transportation, Research Section (ODOT). (2017). Risk factors for pedestrian and bicycle crashes (Publication No. FHWA-OR-RD-17-13; SPR 779). Oregon Department of Transportation. https:/​/​rosap.ntl.bts.gov/​view/​dot/​36343
Poikela, M. E. (2020). Perceived privacy in location-based mobile system. Springer. https:/​/​doi.org/​10.1007/​978-3-030-34171-8
Google Scholar
Queensland Department of Transport and Main Roads (TMR). (2024). Road crash locations. Queensland Government. https:/​/​www.data.qld.gov.au/​dataset/​crash-data-from-queensland-roads
Santa, J., Pereñíguez, F., Moragón, A., & Skarmeta, A. F. (2014). Experimental evaluation of CAM and DENM messaging services in vehicular communications. Transportation Research Part C: Emerging Technologies, 46, 98–120. https:/​/​doi.org/​10.1016/​j.trc.2014.05.006
Google Scholar
Sodnik, J., Dicke, C., Tomažič, S., & Billinghurst, M. (2008). A user study of auditory versus visual interfaces for use while driving. International Journal of Human-Computer Studies, 66(5), 318–332. https:/​/​doi.org/​10.1016/​j.ijhcs.2007.11.001
Google Scholar
Sutton, J., & Wood, M. M. (2025). Opting Out: Over-Alerting and Warning Fatigue in the Era of Wireless Emergency Alerts. Journal of Contingencies and Crisis Management, 33(3), e70076. https:/​/​doi.org/​10.1111/​1468-5973.70076
Google Scholar
Voinea, G. D., Postelnicu, C. C., Duguleana, M., Mogan, G. L., & Socianu, R. (2020). Driving performance and technology acceptance evaluation in real traffic of a smartphone-based driver assistance system. International Journal of Environmental Research and Public Health, 17(19), 7098. https:/​/​doi.org/​10.3390/​ijerph17197098
Google Scholar
Wu, R., & Chen, H. T. (2023). The effect of visual and auditory modality mismatching between distraction and warning on pedestrian street crossing behavior. In 2023 IEEE International Symposium on Mixed and Augmented Reality (ISMAR) (pp. 1045–1054). IEEE. https:/​/​doi.org/​10.1109/​ISMAR59233.2023.00121
Google Scholar
Zhang, C., Wei, J., Qu, S., Huang, C., Dai, J., Fu, P., Wang, Z., & Li, X. (2023). Implementation of a V2P-based VRU warning system with c-v2x technology. IEEE Access, 11, 69903–69915. https:/​/​doi.org/​10.1109/​ACCESS.2023.3293122
Google Scholar
Zoghlami, C., Kacimi, R., & Dhaou, R. (2023). 5G-enabled V2X communications for vulnerable road users safety applications: A review. Wireless Networks, 29(3), 1237–1267. https:/​/​doi.org/​10.1007/​s11276-022-03191-7
Google Scholar

Appendix A. Survey Questions

Software Functions and Alert Methods for Various Road Users Receiving Collision Warnings via Mobi
Question Response options
What is your age group?
(Quit if less than 18 years old)
  • Less than 18
  • 18 - 40
  • 40 - 60
  • Above 60
In your opinion, which places or situations on the road are most prone to collisions or dangers?
(Please select up to 3)
  • Intersections
  • Walking/driving at night
  • Areas with poor visibility
  • High-traffic roads
  • Pedestrians or cyclists not following traffic rules
  • Driver distraction
  • Poor or no traffic signals
  • Sudden appearance or unpredictable movement of motorcycles
  • Others (please specify)
Have you used any road safety applications or devices?
  • Yes (please list)
  • No
How much time do you usually spend driving or cycling each day?
(Please select the option that best describes you)
  • Less than 2 hours per day
  • More than 2 hours per day
  • I don't drive or cycle regularly
Which methods do you prefer for receiving safety warnings?
(Please select up to 3)
  • Mobile phone - sound alerts
  • Mobile phone - vibration alerts
  • Visual displays (e.g., pop-ups on the phone screen)
  • Audio alert through headphones
  • LED light flashing
  • Wearable devices (e.g., smartwatches, bracelets)
  • Visual alert on pedestrian crossing signals
  • Handlebar vibration alert for bicycle
  • Others (please specify)
Which method do you prefer for enabling interaction between vehicles and vulnerable road users (pedestrians and cyclists) to enhance safety?
  • Through short-range direct signals between nearby users (no internet needed)
  • Internet-based connectivity to receive safety warnings (e.g. using apps)
  • A combination of both short-range direct signals and Internet-based methods
  • No interconnection or interaction between road users
  • Other (please specify)
Are you willing to share your location data to enable safety interaction with other road users?
  • Yes, safety is always the most important
  • Yes, if my location privacy can be protected
  • No, I have concerns about my location privacy
  • No, I do not like to interact with other road users
  • None of the above (please specify)
How willing are you to receive safety alerts or warning messages from your smart mobile device (such as a smartphone or tablet) in addition to route guidance?
  • Very willing - I believe it can significantly improve safety
  • Willing - I see the potential benefits and see nothing wrong
  • Neutral - I have no strong opinions either way
  • Somewhat unwilling - I have concerns about distractions and disturbances
  • Very unwilling - I don't see any value in using my device for this purpose
Would you prefer a system that:
  • Issues safety alerts for any potential risks at any time
  • Issues safety alert only for high-risk situations or locations
  • Allows me to adjust how frequently alerts appear (alert frequency setting, including zero alerts)
  • None of the above
Would you prefer a safety alert system in the form of:
  • A standalone safety app (dedicated only to road hazard alerts)
  • Integration with existing navigation apps (e.g., Google Maps, Waze)
  • Both (standalone + integration) - use interchangeably depending on the situation
  • No preference
  • Other (please specify)
If the system triggered frequent "false alarms" in low-risk situations, how likely would you be to continue using it?
  • Very likely, I prefer maximum caution
  • Somewhat likely, I can tolerate occasional false alarms
  • Neutral, uncertain
  • Somewhat unlikely, false alarms reduce my trust significantly
  • Very unlikely, I would probably turn it off
Under what circumstances might you ignore safety warnings?
(Please select up to 3)
  • Warnings are too frequent
  • Focusing on something else
  • Believe the warning is not important
  • The warning method is not noticeable
  • Others (please specify)
If an additional small device is required to support safety interactions, what price range would you find acceptable? (AUD)
  • Less than $300
  • $300 -$500
  • $500 - $1,000
  • More than $1,000
Are you willing to participate in testing use of the small safety devices?
  • Yes (please provide contact)
  • Maybe (please provide contact)
  • No
Which road user roles best describe you on roads?
(Select all that apply)
  • Pedestrian
  • Cyclist
  • Car Driver
If you selected "Pedestrian," please answer the following questions
As a pedestrian, what safety information do you pay most attention to when crossing the street or walking on sidewalks?
(Please select up to 3)
  • Vehicle speed
  • Vehicle distance
  • Turning vehicles
  • Traffic signal status
  • In the blind spots of a vehicle
  • The behaviour of other pedestrians
  • Others (please specify)
Under what circumstances would you find receiving safety warnings most beneficial? (Please select up to 3)
  • When crossing streets or intersections
  • When walking at night
  • In heavy-traffic areas
  • Under conditions of poor visibility (e.g., fog, rain, darkness)
  • In environments with high ambient noise
  • Others (please specify)
When you are in situations where safety warnings are essential (for example, during poor visibility, while crossing roads, or when walking in dark areas), which frequency of safety alerts would you prefer?
  • I prefer to receive frequent safety alerts
  • I prefer to receive alerts at the usual (standard) frequency (e.g. bee bee bee)
  • I prefer to receive fewer alerts and rely on my own judgment
  • I'm not sure
If you selected "Cyclist," please answer the following questions
Which places or road sections feel most dangerous to you when cycling?
(Please select up to 3)
  • Roads with fast-moving vehicles
  • Roads without bicycle lanes
  • Intersections
  • Areas with unclear traffic signals
  • Road sections with poor visibility
  • Others (please specify)
What information would you like to receive when cycling through intersections or crossing roads?
(Please select up to 3)
  • Turning vehicles
  • Vehicles crossing the road
  • Accident-prone area alerts
  • Pedestrians ahead
  • Traffic signal status
  • Others (please specify)
What safety information do you need when sharing the road with vehicles?
(Please select up to 3)
  • Vehicles approaching from behind
  • Vehicles turning left or right
  • Road hazards (potholes, obstacles, etc.)
  • Parked vehicles ahead
  • Vehicles too close to the bicycle
  • Others (please specify)
What do you think is the most appropriate way to issue warnings to cyclists on the roads?
  • Always issue warnings when the other road users are approaching to cyclists within a certain distance
  • Always issue alerts if a fast-running road user is too close to me
  • Issue alerts only when there is a real collision risk
  • Never issue alerts or warnings to avoid false alarms
  • Others (please specify)
If you selected "Car Driver," please answer the following questions
In your driving experience, which of the following situations do you find most hazardous?
(Please select up to 3)
  • Sudden appearance of pedestrians or cyclists
  • Heavy traffic or congestion
  • Adverse weather conditions
  • Nighttime driving
  • Dangerous driving behaviours by other vehicles
  • Others (please specify)
Under which circumstances would receiving safety warnings be most beneficial for you?
(Please select up to 3)
  • When pedestrians or cyclists suddenly enter the roadway
  • When pedestrians or cyclists are in your vehicle's blind spots
  • When pedestrians or cyclists are difficult to see at night
  • When pedestrians or cyclists are not following traffic rules
  • When individuals with disabilities are crossing the road
  • When multiple types of users are sharing the same roads or lanes
  • Others (please specify)
For the safety of drivers and other road users, which of the following approaches would you prefer to use while on the road?
(Please select up to 3)
  • Rely on autonomous vehicles for enhanced safety
  • Utilise dedicated communication networks for safety alerts
  • Receive safety messages on smartphones
  • Drive vehicles equipped with any available connectivity options for safety
  • Others (please specify)
Are your answers primarily based on your road use experiences in Australia or in other countries?
  • In Australia
  • In other countries
  • Both
Do you have any suggestions or comments on improving road safety warning systems? (Please fill in below) [open ended]