Research Abstract
Towards Leveraging Underutilized IoT Resources for Automotive Software: A Study on Resource Sharing for Connected Vehicles
Background: Internet of Things (IoT) has found its way to day-to-day lives
of people, making their lives convenient and connected. As a result of growing
IoT usage, we are surrounded by potentially underutilized computing resources
suggesting an opportunity to improve their utilization by sharing them with
entities that require more resources for complex computations, which may occur
occasionally. However, the resource allocation and resource utilization process
is a rather complex task that involves heterogeneous and distributed entities.
This concept is particularly relevant in highly dynamic and safety-critical
domains such as transportation and automotive systems, where edge devices
such as vehicles operate in a resource-constrained environment. Such entities
could benefit from real-time data processing that could support advanced driver
assistant systems (ADAS) enabling autonomous driving (AD).
Objective: This research focuses on to what extent underutilized resources in
a highly dynamic heterogeneous environment can be utilized more efficiently.
This thesis explores resource sharing in edge devices within close vicinity on
the example of connected vehicles. We demonstrate and evaluate the concept
on a practical application scenario from the automotive domain, where one
vehicle would benefit of being able to “look around the corner”.
Method: This research employs a design science research method to address
the research questions. A systematic mapping study was conducted to identify
the key research areas and limitations of the automotive domain focusing on
Vehicular ad-hoc Networks (VANETs). Based on the results of the systematic
mapping study, an explanatory study was conducted to present the proposed
resource utilization framework exploring the novel research areas identified.
Following that, a series of experimental-based evaluation studies was conducted
to explore the applicability of state-of-the-art LLMs as communication inter-
faces within the proposed resource utilization framework. This line of studies
includes identifying and mitigating the potential challenges of using LLMs as a
tool to support resource utilization.
Findings: The results of the study revealed that resource utilization can be
achieved through sharing underutilized computing resources of nearby IoT-
enabled entities. Within the context of the selected practical application
scenario and the perception-related task, our experiments showed that LLMs
can support pedestrian detection and localization, supporting the concept of
utilizing the computing resources of nearby connected vehicles. LLMs are capa-
ble of initiating a dialogue between connected vehicles and processing relevant
multimodal data to contribute to improved decision-making in autonomous
driving. Further experiments proposed and evaluated novel techniques to
ensure the trustworthiness of such LLM-assisted systems.
Conclusion: The introduction of state-of-the-art AI tools such as LLMs have
the potential to positively impact the perception and monitoring tasks in
ADAS, bringing in a new research dimension to the automotive context. This
novel approach aims to enhance the adaptability and efficiency of the proposed
resource utilization framework for safety critical systems, demonstrating its
capability of addressing the industrially relevant practical application scenario
the let a vehicle “look around the corner”.
Keywords Internet of Things, Resource Utilization, Automotive, Large Lan-
guage Models, Trustworthiness, Hallucination Detection and Mitigation.
