Context-Aware Warning Systems: Leveraging Driving Environment

Data for Improved Driver and Road User Warnings

Alexander Stocker

, Tahir Emre Kalayci

, Michael Spitzer

and Gerald Musser

Virtual Vehicle Research GmbH, Inffeldgasse 21a, 8010 Graz, Austria

Keywords: Software-Defined Vehicles, Connected Vehicles, Driver Warning, Driver Monitoring, Data Sharing, Data

Ecosystems.

Abstract: Web technologies, Internet of Things (IoT) frameworks, and modern communication standards are

increasingly transforming the automotive sector, giving rise to software-defined vehicles. These vehicles

operate as connected entities within a broader digital ecosystem, enabling real-time data exchange with

infrastructure, cloud services, and other road users. This ongoing digitalization opens new opportunities to

improve road safety through intelligent, context-aware driver assistance systems. Our paper introduces a novel

context-aware driver warning system to be developed as part of the ROADGUARD project. The system will

fuse data from in-cabin driver monitoring with data about the external driving environment to enhance the

accuracy and contextual relevance of safety alerts. Conventional Driver Monitoring Systems (DMS) often

rely solely on gaze-based heuristics, which can lead to false positives when environmental context is not

considered. Our approach will overcome this limitation by integrating multimodal sensing, AI-driven edge

inference, secure data sharing, and adaptive, multi-target warning delivery. Our proposed system architecture

is structured around three interconnected subsystems—Sensing, Sharing, and Acting. It will not only enable

more precise, real-time alerts for drivers but also cooperative warnings for vulnerable road users such as

pedestrians and cyclists. By embedding situational awareness and supporting data-driven improvement via

mobility data spaces, our system supports the Vision Zero objective of eliminating traffic fatalities.

1 INTRODUCTION AND

MOTIVATION

Web technologies, Internet of Things (IoT)

frameworks, and modern communication standards

are increasingly permeating the automotive domain

(Sterk et al. 2023; Kaiser et al., 2021). This

technological convergence is transforming vehicles

into connected entities within a broader digital

ecosystem, enabling real-time data exchange with

cloud services, roadside infrastructure, and other

vehicles (Stocker et al., 2017). Simultaneously,

vehicles are evolving into software-defined platforms

(Liu et al. 2022), moving beyond traditional

hardware-centric architectures. This shift toward the

software-defined vehicle (Otto et al. 2025) offers

greater flexibility, supports over-the-air (OTA)

https://orcid.org/0000-0002-3758-1617

https://orcid.org/0000-0001-6228-1221

https://orcid.org/0000-0003-2173-9317

https://orcid.org/0009-0005-6006-588X

updates, and facilitates the integration of advanced

functionalities such as driving automation, context-

aware driver assistance, continuous system

optimization, and intelligent decision-making.

Moreover, vehicles are now enabled to leverage a

wide range of external data sources (Kaiser et al.

2017), such as real-time traffic information, weather

reports, and transportation infrastructure data, to

further enhance both safety and driver comfort. By

integrating these heterogeneous inputs, vehicles

become capable of making context-sensitive

decisions (Matalonga et al., 2025), anticipating

potential hazards, tailoring warnings to situational

risks, and delivering more accurate and timely

assistance to drivers (Ebinger et al., 2024).

From a digitalization standpoint, product

architectures are undergoing a fundamental

Stocker, A., Kalayci, T. E., Spitzer, M. and Musser, G.

Context-Aware Warning Systems: Leveraging Driving Environment Data for Improved Driver and Road User Warnings.

DOI: 10.5220/0013820600003985

Paper published under CC license (CC BY-NC-ND 4.0)

In Proceedings of the 21st International Conference on Web Information Systems and Technologies (WEBIST 2025), pages 573-581

ISBN: 978-989-758-772-6; ISSN: 2184-3252

573

transformation (Yoo et al., 2024). Traditional vehicle

development, which focused primarily on mechanical

systems and static design paradigms, is giving way to

flexible, modular, and software-centric architectures

and services (Otto et al. 2025). In this new paradigm,

digital innovation, service-oriented design, and

platform-based thinking are becoming central (Sterk

et al., 2024). As vehicles grow increasingly connected

and data-driven (Stocker et al., 2017), value creation

is shifting from physical products to digital

services—such as real-time diagnostics, predictive

maintenance, personalized driver assistance, and

subscription-based feature upgrades (Stocker et al.

2021). This evolution not only transforms how

vehicles are designed and maintained but also

redefines how users interact with and experience

mobility.

This digital transformation of the mobility

automotive sectors (Stocker et al., 2024; Piccini et al.

2015) holds significant potential to improve one of

the most critical aspects of mobility, driving and road

safety. By harnessing connected technologies, real-

time data, and intelligent systems, vehicles can for

instance better detect, predict, and respond to

hazardous situations—ultimately reducing the risk of

accidents and enhancing protection for both drivers

and vulnerable road users.

Inattentiveness and distraction remain among the

leading contributors to road accidents globally

(Regan et al. 2011). In response, and to move closer

to the European Union’s Vision Zero objective

(European Commission, 2022), aiming to eliminate

all traffic-related fatalities, modern vehicles are

increasingly equipped with driver assistance and

warning systems (Li et al. 2024) designed to detect

and mitigate risks in real-time. However, despite

these advancements, the persistent rate of road

fatalities—especially among pedestrians, cyclists,

and motorcyclists (Silla et al. ,2017)—highlights the

urgent need for more effective and intelligent safety

interventions. Consequently, Physical AI is

recognized as a pivotal enabler of digital innovation

(Stocker, 2025).

One key limitation of many current driver

assistance, warning systems and driver warning apps

(Trager et al. 2021) is their tendency to generate false

positives (Large et al. 2017), alerts triggered in non-

critical situations. These frequent, often unnecessary

warnings can lead to driver desensitization, reduced

trust in the system, and diminished overall safety

benefits (Navarro et al. 2016).

To increase the relevance and effectiveness of in-

vehicle driver warning systems, we propose the

integration of data and information from both the

external driving environment and in-cabin sources,

such as driver state and behaviour. By fusing

environmental context—such as road conditions,

traffic dynamics, and weather—with insights from

driver monitoring systems (e.g., gaze tracking,

drowsiness detection, and cognitive load estimation),

warning systems can become more adaptive, timely,

and personalized. Our multimodal fusion approach

has the potential to significantly reduce false

positives, increase contextual relevance, and improve

driver trust and responsiveness to safety-critical

alerts.

Based on the introduction and motivation, Section

2 presents our proposed context-aware warning

system for drivers and other road users, situated

within the current state of the art in driver warning

systems. This section provides a detailed overview of

the motivating scenario and the rationale behind our

solution concept. Section 3 outlines the solution

architecture and its technical foundation, organized

around the three core subsystems: Sensing, Sharing,

and Acting. Section 4 discusses our approach along

with key insights from technological considerations.

Finally, Section 5 concludes the paper with a

summary and an outlook on future research directions

and potential applications.

2 STATE OF THE ART

AND ADVANCEMENTS

2.1 Driver Monitoring and Warning

Systems

Current Driver Monitoring Systems (DMS) primarily

focus on observing the driver within the cabin

(Michelaraki et al., 2023), often neglecting the

broader external driving context. Moreover, existing

digital warning strategies are typically designed to

alert the driver alone, overlooking the potential

benefit of also communicating risks to nearby road

users—especially when the vehicle is operated by a

distracted or inattentive driver.

A key limitation of many current inattention and

distraction detection approaches is their susceptibility

to false positives (cf. Large et al., 2017). These

systems may flag off-road gaze behaviour as

distraction, even when such behaviour is contextually

appropriate, for instance, when a driver is monitoring

a child pedestrian near the roadway. Such

misclassifications not only reduce the relevance of

warnings but can also frustrate drivers, eroding trust

in the system. As a result, drivers may choose to

WEBIST 2025 - 21st International Conference on Web Information Systems and Technologies

574

Figure 1: An example scenario that highlights the limitations of conventional systems.

disable these safety features altogether, ultimately

compromising their effectiveness.

Furthermore, most DMS architectures treat

inattention and distraction as an isolated, driver-

centric problem. They rarely consider how driver

inattention and distraction might impact other road

users in the surrounding environment, nor do they

issue proactive alerts to other road users and

especially to vulnerable road users who may be at risk

due to an inattentive driver. This lack of context-

awareness and bidirectional warning mechanisms

limits the broader safety potential of current driver

monitoring and warning systems.

2.2 Scenario: Context Aware Driver

Monitoring

Before outlining our proposed approach, we begin

with a user scenario that highlights the limitations of

conventional camera-based driver monitor warning

systems (Figure 1). These systems often struggle to

integrate the detected driver state with a dynamic,

contextual understanding of the external

environment, which can lead to delayed,

inappropriate, or misleading responses.

Imagine a vehicle navigating an urban

environment. A camera-based Driver Monitoring

System (DMS) tracks the driver's gaze and eye

movements, issuing warnings based on simple

heuristics—for example, triggering an alert if the

driver looks away from the road for an extended

period (see Figure 1, upper box).

As an example, the China NCAP safety rating for

driver attention monitoring systems mandates issuing

escalation alerts if the driver’s eyes are off the road

for more than 3 to 5 seconds (Medium, 2025). For

Europe, Euro NCAP defines a long distraction as a

single instance of the driver’s gaze away from the

forward road lasting between 3 and 4 seconds (Euro

NCAP, 2025).

Now consider the following situation: A driver

turns the gaze to the left and keeps watching a dog

and a child playing near the curb. The driver is

rightfully cautious and remains prepared to brake, as

this is an exceptional case where the general trust

principle does not apply. Although the driver is acting

safely and attentively, the system may still trigger a

distraction warning because it lacks awareness of the

context—specifically, that the off-road gaze is

purposeful and safety-motivated (Figure 1, lower left

box). Such false positives can frustrate the driver and

lead to disengagement or deactivation of the warning

system.

Now imagine that, simultaneously, a pedestrian

Context-Aware Warning Systems: Leveraging Driving Environment Data for Improved Driver and Road User Warnings

575

Figure 2:

The proposed

context-aware driver monitoring and warning

solution.

begins crossing the street from the right—outside

the driver’s current field of view, as their attention

remains focused on the child and dog. The driver fails

to notice the pedestrian, significantly increasing the

risk of a potentially fatal collision. Crucially, the

current Driver Monitoring System (DMS) would not

issue any alert in this situation because it lacks

situational awareness of the external environment and

the evolving traffic scenario (Figure 1, lower right

box). Additionally, the pedestrian is not warned that

the driver’s attention is elsewhere, leaving them

vulnerable. Finally, other road users remain unaware,

too, when a driver’s attention is diverted away from

the road.

3 SOLUTION APPROACH

Our proposed solution to be developed in the

roadguard project

is structured around three core

subsystems essential for context-aware driver

monitoring and warning: Sensing, Sharing, and

Acting (Figure 2). Together, these components form

an integrated framework for comprehensive safety

monitoring that addresses both the driver and

surrounding road users. At the heart of this

architecture are AI-enabled edge devices, which

enable real-time analysis and decision-making

directly within the vehicle.

Roadguard project description: https://projekte.ffg.at/

projekt/5126443

The Sensing subsystem captures data from both

inside and outside the vehicle, including driver state,

environmental context, and potential hazards, thereby

handling manifold data integration challenges on the

vehicle edge computer (cf. Kalayci et. al. 2021). The

Sharing subsystem manages the secure transmission

of relevant data across vehicles and infrastructure,

with a strong emphasis on data privacy and regulatory

compliance. The Acting subsystem plays a pivotal

role in delivering context-sensitive warnings, not only

to the driver but also to nearby vulnerable road users,

thereby enhancing situational awareness for all

parties.

Our tripartite framework aims to address the

shortcomings of current driver warning systems by

ensuring that relevant information is captured,

securely exchanged, and acted upon in a timely and

context-aware manner. The ongoing digitalization of

the automotive sector—coupled with advancements

in web technologies, IoT frameworks, and

standardized communication protocols—provides a

robust foundation for implementing such an

integrated system.

3.1 Sensing System: Contextual Data

Acquisition and Processing

The sensing system at the core of our approach

consists of two complementary subsystems designed

WEBIST 2025 - 21st International Conference on Web Information Systems and Technologies

576

to monitor both the driver’s state and the external

driving environment. This dual-sensing strategy

provides a comprehensive and context-aware

understanding of risk, which is crucial for delivering

effective and timely warnings to drivers and other

road users.

The in-cabin sensing unit focuses on assessing the

driver’s attention and cognitive state. It incorporates

dual near-infrared (NIR) stereo cameras positioned to

unobtrusively monitor the driver’s head and eye

movements. Computer vision algorithms process the

stereo video streams to perform gaze estimation,

blink detection, head pose tracking, and fatigue

analysis. Driver state classification algorithms then

generate outputs such as attentive versus inattentive,

drowsy versus alert, and gaze-on-road versus gaze-

off-road indicators.

An external sensing unit captures the driving

context and detects potential hazards. This subsystem

includes RGB cameras and utilizes object detection

(cf. Kalayci et al. 2022) and tracking models to

identify vulnerable road users (VRUs), other

vehicles, and static obstacles in real time.

Additionally, trajectory prediction models forecast

the likely motion of detected road users, enabling the

system to anticipate emerging threats.

All sensor data, both internal and external, are

processed and fused on a dedicated in-vehicle edge

device equipped with AI inference capabilities. This

local processing unit enables real-time decision-

making under strict latency constraints. It evaluates

whether the driver’s current behaviour is safe within

the given environmental context and determines

whether a warning should be issued to the driver, to

nearby or mid-distance road users, or if data should

be securely shared externally with connected

infrastructure or vehicles, in full compliance with

data privacy standards.

3.2 Sharing System: Secure Data

Exchange and Ecosystem

Integration

Selected data, such as instances of critical driver

states or high-risk scenarios, are transmitted by the

Sharing subsystem through secure communication

channels to a centralized server infrastructure. This

transmission process adheres to data protection

regulations and ensures both the confidentiality and

integrity of sensitive information.

Once received, the data on the driver state follows

two primary pathways: integration into a training data

repository and contribution to other data spaces. A

subset of the collected data is anonymized and stored

in a dedicated training dataset to support the ongoing

enhancement of machine learning models for driver

state monitoring as well as environmental perception

and prediction. By incorporating real-world examples

into the training pipeline, the system enables adaptive

learning and continuous performance refinement.

Figure 3:

The proposed high-level

architecture

of the solution.

Context-Aware Warning Systems: Leveraging Driving Environment Data for Improved Driver and Road User Warnings

577

In parallel, relevant portions of the data are shared

with broader mobility data ecosystems using

standardized protocols and interoperable connectors

(e.g., GAIA-X, IDSA). This facilitates data exchange

with external stakeholders such as traffic

management authorities, urban planners, and fleet

operators—enabling coordinated intelligence and the

development of advanced mobility services.

This dual-purpose data sharing approach

simultaneously supports real-time safety

interventions and long-term system advancement

through collaborative learning and ecosystem-wide

integration.

3.3 Acting System: Context-Aware

Warning Presentation

The Acting subsystem is responsible for issuing

context-sensitive warnings based on the combined

analysis of driver state and external environmental

data. It differentiates between warnings intended for

the ego vehicle—namely the driver and in-vehicle

systems—and those aimed at external road users in

close or mid-range proximity.

Warnings for the driver are delivered through a

dedicated user interface implemented as an

application running on the Android Automotive

Operating System (AAOS) (cf. Sterk et al., 2023).

This interface receives warning signals from the

sensing subsystem and provides real-time alerts using

visual cues such as screen overlays, ambient lighting,

or flashing icons, and auditory feedback such as tones

or voice prompts. This configuration ensures that the

driver receives immediate and relevant notifications

in situations involving distraction, inattention, or

emerging hazards.

For vulnerable road users in the immediate

vicinity of the vehicle, the system uses lighting-based

external warning signals. These can include flashing

headlights or taillights, projection-based indicators,

or dynamic lighting effects that visually communicate

a distracted driver state or imminent danger. These

warnings are designed to increase the situational

awareness of pedestrians and cyclists, particularly

when the driver is not fully attentive.

For road users at mid-range distances—such as

nearby vehicles or motorcycles—cooperative

warnings are issued via connected applications. In

vehicles equipped with AAOS, a connected warning

app receives relevant driver state information from

the sharing subsystem and displays appropriate alerts.

For motorcycles and other vehicles not equipped with

AAOS, a standalone Android mobile application is

under development to receive and visualize warnings.

If the data-sharing mechanism supports sufficiently

low-latency communication, both in-vehicle and

environmental perceptions - such as distractions

caused by other road users - can be shared in near

real-time to provide warnings to approaching drivers.

This approach extends the system’s safety reach

beyond the immediate surroundings of the vehicle

and supports broader cooperative awareness, which is

especially valuable in urban and mixed-traffic

environments.

4 TECHNICAL

CONSIDERATIONS

This position paper proposes a context-aware driver

monitoring system that integrates in-cabin data with

environmental information to deliver more effective

and relevant driver warnings. Unlike conventional

driver monitoring systems, which often suffer from

false-positive rates due to focusing on in-cabin

information only and neglecting external context, the

proposed approach fuses multiple data sources and

leverages AI-driven edge computing to enhance

decision-making. By combining insights from driver

behavior with real-time environmental perception,

the system not only improves the accuracy and

timeliness of driver alerts but also extends safety

benefits to vulnerable road users through cooperative

warnings. This context-aware design shows strong

potential to advance road safety, reduce distractions

from unnecessary alerts, and create a more proactive

and adaptive warning frmework.

However, several challenges remain. Ensuring

robust and reliable sensing under diverse real-world

conditions—such as adverse weather or complex

urban environments—requires ongoing refinement of

perception algorithms and models. Advanced AI

frameworks play a crucial role in this process by

enabling adaptive learning, sensor fusion, and

context-aware interpretation of multimodal data.

Leveraging techniques such as deep learning and

transfer learning allows the system to improve over

time by incorporating new data and handling edge

cases more effectively. Moreover, implementing AI

models on resource-constrained edge devices

demands efficient architectures and optimization

strategies to balance accuracy with real-time

performance and power consumption. Addressing

these challenges is essential to maintain high

detection accuracy, reduce false positives, and ensure

the system’s resilience in diverse driving scenarios.

WEBIST 2025 - 21st International Conference on Web Information Systems and Technologies

578

In our sharing subsystem, choosing between

streaming services and simple REST interfaces

involves important trade-offs related to data

processing and fusion capabilities. Streaming

services typically require significant preprocessing

and formatting of data before transmission. This

upfront processing can limit flexibility downstream,

as the data arrives in a more fixed, aggregated form,

reducing the ability to perform on-the-fly merging or

complex fusion with other data sources at the

receiver’s end. On the other hand, REST interfaces,

where data is requested and received in discrete

chunks, allow for more dynamic handling. Since data

arrives in smaller, more raw pieces, it enables real-

time or on-demand computations and merging at the

receiver. This approach preserves the flexibility to

adapt processing strategies based on current context,

system state, or additional inputs, improving the

overall responsiveness and accuracy of fused

information. Therefore, while streaming services can

be beneficial for continuous high-throughput data

flows, they may constrain adaptability due to early-

stage preprocessing. RESTful approaches offer

greater flexibility for on-the-fly calculations and data

fusion but may introduce some latency due to the

request-response nature. Selecting the appropriate

method depends on the specific requirements for

latency, processing load, and fusion complexity in the

sharing subsystem.

As a first step and proof of concept, a simple

REST-based approach will be implemented. In this

setup, compressed messages containing basic

information—such as hazard type, time and

position—will be transmitted to the sharing

component. Receivers can retrieve these messages

based on a self-defined area of interest, allowing for

flexible control over update frequency and the

volume of data processed at once. Additionally, the

sharing backend will automatically filter out outdated

information, reducing both data transfer and

computational load on the receiver side. In a

production-grade system, a hybrid approach

combining streaming, on-demand requests, and

potentially locally cached historical data on the

receiver side could offer an effective solution. This

setup would minimize data transfer, enable fast

message processing, and maintain flexibility in

selecting the type of data and update frequency

according to the receiver’s needs.

The Acting subsystem connects the system’s

insights to real-world safety actions by delivering

context-aware warnings to drivers and nearby road

users. Leveraging IoT principles and web

technologies, it enables seamless communication and

interaction across vehicles and devices. Integration

with Android Automotive OS (AAOS) allows the

system to deploy flexible, app-based warnings inside

the vehicle using native features like visual alerts and

sounds. Externally, the subsystem uses connected

lighting systems and projection-based signals to warn

vulnerable road users such as pedestrians and cyclists

in real time. Additionally, via secure web-based APIs

and standardized data-sharing protocols, mid-range

warnings are pushed to other vehicles or mobile

apps—extending situational awareness through

cooperative communication within an IoT-enabled

mobility ecosystem. This combination of modern

web standards, edge computing, and IoT connectivity

ensures adaptive, reliable, and comprehensive

warning delivery that enhances safety both inside and

beyond the vehicle.

Privacy and data security are critical

considerations, especially given the sensitive nature

of driver state data and the need for secure data

sharing across vehicles and infrastructure. User

acceptance and trust also hinge on minimizing

intrusive or excessive warnings, necessitating careful

calibration of system sensitivity and thoughtful

human-machine interface design.

A central aspect of our system design is strict

compliance with data protection and privacy

regulations, reflecting the dual nature of sensing both

inside and outside the vehicle. In-cabin driver

monitoring involves sensitive personal data, such as

driver images, which require explicit user consent and

stringent handling procedures. To address this,

driver-related data is processed locally on the

vehicle’s edge device and remains within the vehicle

unless lawful data sharing with explicit consent

occurs as managed by the secure data transfer

module, thus minimizing data exposure.

Conversely, external environmental monitoring

captures anonymized, non-identifiable information

about road users and hazards in public spaces, which

generally does not require individual consent but

follows privacy-by-design principles. The overall

architecture enforces stringent data governance,

ensuring personal data does not leave the vehicle

without proper safeguards, while external sensing

enhances safety without infringing on individual

privacy rights.

5 CONCLUSION

Our concept of a context-aware driver warning

system demonstrates how integrating in-cabin

monitoring with environmental sensing, enabled by

Context-Aware Warning Systems: Leveraging Driving Environment Data for Improved Driver and Road User Warnings

579

IoT and Web technologies, can significantly enhance

road safety. By leveraging edge AI and secure data

sharing, our system will deliver timely, relevant alerts

to drivers and other road users alike. Our tripartite

framework—Sensing, Sharing, and Acting—

addresses key limitations of conventional approaches,

reducing false positives and expanding cooperative

awareness.

ACKNOWLEDGEMENTS

The research presented in this paper was funded by

the Österreichische Forschungsförderungs-

gesellschaft mbH (FFG) under the “Digital Road User

Safeguarding: A Regulatory-Compliant, Contextual

Driver State Assessment & Road User Warning

System (ROADGUARD)” project.

REFERENCES

Ebinger, N., Neuhuber, N., Moser, J., Trösterer, S., &

Stocker, A. (2024). Which partially automated driving

function do drivers prefer? Results from two field

studies on public highways. Transportation

Engineering, 17, 100236.

European Commission (2022). EU Road Safety: Towards

"Vision Zero". https://cinea.ec.europa.eu/publications/

digital-publications/eu-road-safety-towards-vision-

zero_en

Euro NCAP, 2025. Safe Driving. Driver Engagement.

Protocol.

https://www.euroncap.com/media/85854/euro-ncap-

protocol-safe-driving-driver-engagement-v10.pdf

Kaiser, C., Festl, A., Pucher, G., Fellmann, M., & Stocker,

A. (2019). The Vehicle Data Value Chain as a

Lightweight Model to Describe Digital Vehicle

Services. In WEBIST (pp. 68-79).

Kaiser, C., Stocker, A., Viscusi, G., Fellmann, M., &

Richter, A. (2021). Conceptualising value creation in

data-driven services: The case of vehicle data.

International Journal of Information Management, 59,

102335.

Kalayci, T. E., Kalayci, E. G., Lechner, G., Neuhuber, N.,

Spitzer, M., Westermeier, E., & Stocker, A. (2021).

Triangulated investigation of trust in automated

driving: Challenges and solution approaches for data

integration. Journal of Industrial Information

Integration, 21, 100186.

Kalayci, T. E., Ozegovic, G., Bricelj, B., Lah, M., &

Stocker, A. (2022). Object detection in driving datasets

using a high-performance computing platform: A

benchmark study. IEEE Access, 10, 61666-61677.

Large, D. R., Harvey, C., Burnett, G., Merenda, C., Leong,

S., & Gabbard, J. (2017). Exploring the relationship

between false alarms and driver acceptance of a

pedestrian alert system during simulated driving.

Liu, Z., Zhang, W., & Zhao, F. (2022). Impact, challenges

and prospect of software-defined vehicles. Automotive

Innovation, 5(2), 180-194.

Li, C., Xu, A., Sachdeva, E., Misu, T., & Dariush, B. (2024,

May). Optimal Driver Warning Generation in Dynamic

Driving Environment. In 2024 IEEE International

Conference on Robotics and Automation (ICRA) (pp.

14184-14190). IEEE.

Matalonga, S., Amalfitano, D., Solari, M., Hauck, J. C. R.,

& Travassos, G. H. (2025). Testing Context-Aware

Software Systems From the Voices of the Automotive

Industry. IEEE Transactions on Industrial Informatics.

Medium (2025), China NCAP Explained: New Safety

Ratings for Driver Monitoring and In-Cabin Systems,

May 14, 2025.

Michelaraki, E., Katrakazas, C., Kaiser, S., Brijs, T., &

Yannis, G. (2023). Real-time monitoring of driver

distraction: State-of-the-art and future insights.

Accident Analysis & Prevention, 192, 107241.

Navarro, J., Yousfi, E., Deniel, J., Jallais, C., Bueno, M., &

Fort, A. (2016). The impact of false warnings on partial

and full lane departure warnings effectiveness and

acceptance in car driving. Ergonomics, 59(12), 1553-

1564.

Otto, S., Wlcek, M., & Wortmann, F. (2025). Towards

Conceptualizing Software-Defined Vehicles: A

Systematic Review and Future Research Avenues.

Piccinini, E., Hanelt, A., Gregory, R., & Kolbe, L. (2015).

Transforming industrial business: The impact of digital

transformation on automotive organizations.

International Conference on Information Systems

(ICIS).

Regan, M. A., Hallett, C., & Gordon, C. P. (2011). Driver

distraction and driver inattention: Definition,

relationship and taxonomy. Accident Analysis &

Prevention, 43(5), 1771-1781.

Silla, A., Rämä, P., Leden, L., Van Noort, M., de Kruijff,

J., Bell, D., Morris, A., Hancox, G., and Scholliers, J.

(2017). Quantifying the effectiveness of ITS in

improving safety of VRUs. IET Intelligent Transport

Systems, 11(3), 164-172.

Sterk, F., Stocker, A., Heinz, D., & Weinhardt, C. (2024).

Unlocking the value from car data: A taxonomy and

archetypes of connected car business models.

Electronic Markets, 34(1), 13.

Sterk, F., Heinz, D., Hengstler, P., & Weinhardt, C. (2023).

Reallocating uncertainty in incumbent firms through

digital platforms: The case of Google’s automotive

ecosystem involvement. ICIS 2023 Proceedings, 1, 16.

Stocker, A., Kaiser, C., & Fellmann, M. (2017). Quantified

vehicles: Novel services for vehicle lifecycle data.

Business & information systems engineering, 59, 125-

130.

Stocker, A., Lechner, G., Kaiser, C., & Fellmann, M.

(2021). Digitalized Mobility. In Americas Conference

on Information Systems (AMCIS).

WEBIST 2025 - 21st International Conference on Web Information Systems and Technologies

580

Stocker, A., Kaiser, C., Lechner, G., & Fellmann, M.

(2024). A conceptual framework for mobility data

science. IEEE Access.

Stocker, A. (2025). User Archetypes of Physical AI

Systems: Insights from an Automated Driving Field

Study, European Conference on Information Systems -

ECIS.

Trager, J., Kalová, L., Pagany, R., & Dorner, W. (2021).

Warning apps for road safety: A technological and

economical perspective for autonomous driving–the

warning task in the transition from human driver to

automated driving. International Journal of Human–

Computer Interaction, 37(4), 363-377.

Yoo, Y., Henfridsson, O., Kallinikos, J., Gregory, R.,

Burtch, G., Chatterjee, S., & Sarker, S. (2024). The next

frontiers of digital innovation research. Information

Systems Research, 35(4), 1507-1523.

Context-Aware Warning Systems: Leveraging Driving Environment Data for Improved Driver and Road User Warnings

581