Advances in intelligent vehicular health monitoring and fault diagnosis: Techniques, technologies, and future directions
Despite numerous approaches tailored for specific components and vehicle types, the automotive industry lacks a unified framework that integrates diverse methodologies while addressing scalability, adaptability, and efficiency across conventional (CVs), electric (EVs), and autonomous vehicles (AVs)....
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier Ltd
2025
|
| Subjects: | |
| Online Access: | http://umpir.ump.edu.my/id/eprint/44533/ http://umpir.ump.edu.my/id/eprint/44533/1/Advances%20in%20intelligent%20vehicular%20health%20monitoring%20and%20fault%20diagnosis.pdf |
| Summary: | Despite numerous approaches tailored for specific components and vehicle types, the automotive industry lacks a unified framework that integrates diverse methodologies while addressing scalability, adaptability, and efficiency across conventional (CVs), electric (EVs), and autonomous vehicles (AVs). This study provides a taxonomic classification of vehicle health monitoring (VHM) and fault diagnosis techniques, highlighting the transformative role of advanced sensors, IoT integration, artificial intelligence (AI), and multi-sensor fusion. Through comparative evaluation of model-based, signal-based, and data-driven approaches, we identify their strengths, limitations, and optimal applications in different vehicular contexts supported by quantitative performance metrics. Key findings demonstrate that AI-driven fault diagnosis in CVs achieves up to 95 % accuracy in early mechanical failure detection, reducing maintenance costs and preventing catastrophic failures. Advanced battery monitoring techniques for EVs improve energy efficiency (15–20 %) and extend battery lifespan (10–15 %), addressing critical range anxiety and sustainability concerns. In AVs, sensor fusion and AI-based prediction achieve 99% reliability in real-time decision-making, enhancing operational safety and passenger confidence. We propose a novel conceptual framework integrating sensors, IoT, AI, and big data analytics to enhance VHM capabilities across all vehicle types. This framework addresses challenges in heterogeneous data integration, efficiency, and real-time processing in dynamic automotive environments. Future research addresses cybersecurity vulnerabilities, optimising AI models for edge computing deployments, developing AI techniques for transparent diagnostics, and leveraging vehicle-to-vehicle communication protocols to improve fault detection accuracy. This study contributes to the reactive to predictive maintenance strategies, potentially transforming maintenance practices across the transportation sector while enhancing vehicle reliability, safety, and sustainability. |
|---|