INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,  
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)  
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XIV, Issue XII, December 2025  
Predictive Health Monitoring Systems for Electric Vehicle  
Powertrains Using Edge AI and CAN Bus Data  
Dhage Abhishek Yuvraj, Giri Aniket Atresh, Prof. Urmila Burde  
Department of Electronics and Telecommunication Engineering, Ajeenkya DY Patil School of  
Engineering, Pune, India  
DOI : https://doi.org/10.51583/IJLTEMAS.2025.1412000134  
Received: 23 December 2025; Accepted: 27 December 2025; Published: 17 January 2026  
ABSTRACT  
Electric vehicles (EVs) are becoming increasingly important in the shift toward sustainable mobility. While their  
adoption is accelerating, ensuring the health and reliability of EV powertrains remains a critical challenge.  
Failures in subsystems such as batteries, motors, and controllers may cause unexpected breakdowns, reduced  
efficiency, and safety issues. Predictive Health Monitoring (PHM) systems aim to prevent such failures by  
identifying anomalies before they escalate. Traditional PHM solutions often depend on cloud platforms, but  
these face drawbacks such as latency, high bandwidth requirements, and privacy risks. To address these  
challenges, edge-based PHM employs embedded devices to process Controller Area Network (CAN) bus data  
locally, enabling real-time diagnostics, enhanced privacy, and cost efficiency.  
This paper presents a structured survey of PHM techniques for EV powertrains using Edge Artificial Intelligence  
(Edge AI) and CAN bus data. The contributions include: (i) classification of PHM approaches into model-based,  
data-driven, security- focused, and edge-deployed methods, (ii) comparative analysis of recent works, (iii) a  
summary table of surveyed papers, (iv) discussion of hardware implementations reported in literature, and (v)  
identification of future research directions.  
Key Words - Electric Vehicles, Predictive Maintenance, Edge AI, CAN Bus, Vehicle Diagnostics, Anomaly  
Detection.  
INTRODUCTION  
The transportation sector is witnessing a global transition toward electric mobility. EVs are valued for their  
efficiency, reduced environmental impact, and compliance with stricter emission norms. At the heart of every  
EV lies the powertrain, consisting of batteries, motors, controllers, and power electronics. Maintaining the health  
of these systems is crucial for safety and performance.  
Traditional diagnostic systems are reactive in nature, identifying issues only after they occur. Cloud-based  
predictive approaches improve fault anticipation but face challenges such as reliance on internet connectivity,  
increased latency, and data privacy concerns. Edge Artificial Intelligence (Edge AI) provides an alternative by  
processing vehicle data locally using embedded devices, enabling real-time fault detection while minimizing  
external dependencies.  
This survey reviews recent advances in predictive health monitoring for EV powertrains using Edge AI and CAN  
bus data. The contributions of this paper are as follows:  
1. A detailed classification of PHM research into four categories.  
2. A comparative review of literature from 2020–2025.  
3. Inclusion of hardware-oriented implementations from recent studies.  
www.ijltemas.in  
Page 1534  
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,  
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)  
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XIV, Issue XII, December 2025  
4. Identification of open challenges and research opportunities for the future.  
BACKGROUND  
Controller Area Network (CAN Bus)  
The CAN bus is a standard in-vehicle communication protocol that enables data exchange between electronic  
control units (ECUs), battery management systems, motor controllers, and sensors. It serves as the primary data  
source for PHM, offering real-time information on voltage, current, temperature, and system faults.  
Edge Artificial Intelligence (Edge AI)  
Edge AI refers to running machine learning models on embedded devices instead of cloud servers. This reduces  
latency, lowers dependence on internet connectivity, and safeguards privacy. Platforms such as Raspberry Pi  
and ESP32, when combined with optimized libraries like TensorFlow Lite, can perform anomaly detection on  
CAN data in real time, making them suitable for predictive monitoring applications.  
LITERATURE SURVEY  
I. Patil et al. (2025) introduced an AI-based predictive maintenance framework that combines Long Short-  
Term Memory (LSTM) networks with Federated Learning for secure fault detection in EVs [1]. While the  
approach enhances privacy and supports predictive modelling, it imposes significant computational  
requirements, limiting deployment on embedded devices.  
II.  
In another study, Wang et al. (2025) utilized Support Vector Machines (SVM) and K-Means clustering on  
On-Board Diagnostics (OBD-II) data to monitor emissions and driving behaviour [2]. The method enables  
improved vehicle diagnostics but requires standardized OBD-II datasets for consistent integration with CAN bus  
analytics.  
III.  
Ahmed and Patel (2025) explored the use of Quantum Artificial Intelligence in combination with Federated  
Learning for EV fault diagnosis [3]. Their method achieved high accuracy in detecting system faults; however,  
it depends on experimental quantum hardware, which is not yet widely available for practical deployment.  
IV.  
Roy et al. (2025) proposed a privacy-preserving predictive maintenance system using Homomorphic  
Encryption integrated with Edge AI [4]. This technique ensures secure CAN diagnostics by safeguarding  
sensitive data, but the computational overhead places a heavy load on edge devices.  
V. Zhang et al. (2024) conducted a survey of anomaly detection methods in in-vehicle networks, focusing on  
deep learning techniques and CAN bus analysis [5]. While their review provides a strong foundation for machine  
learning-based diagnostics, it lacks real- world validation.  
VI.  
Lee et al. (2024) investigated battery health prediction using Deep Belief Networks (DBN) and Recurrent  
Neural Networks (RNN) [6]. Their model achieved accurate State of Health (SoH) estimation but remains  
unsuitable for resource-constrained edge devices, highlighting the need for lightweight alternatives.  
VII.  
Singh et al. (2024) adopted Reinforcement Learning to optimize energy management in hybrid vehicles  
[7]. Although the approach is not explicitly health-focused, it shows potential for enhancing motor control fault  
tolerance.  
VIII.  
Fernandez et al. (2024) presented a digital twin framework to simulate EV powertrain operations and  
evaluate faults virtually [8]. The framework enables safe prototyping and model validation but requires a  
complex and resource-intensive setup.  
IX.  
Gupta and Sharma (2023) employed statistical models and machine learning techniques to estimate the  
Remaining Useful Life (RUL) of automotive components [9]. While their models align with industry standards,  
www.ijltemas.in  
Page 1535  
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,  
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)  
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XIV, Issue XII, December 2025  
they face challenges in real-time applications.  
X. Hossain et al. (2023) applied Convolutional Neural Networks (CNN), SVM, and time-series analysis for  
lifecycle-based predictive maintenance in industrial equipment [10]. Although not EV-specific, the approach  
demonstrates transferable concepts applicable to EV systems.  
XI.  
Kumar and Singh (2023) reviewed cybersecurity measures for electric vehicles, focusing on CAN bus  
security protocols and Intrusion Detection Systems (IDS) [11]. While their study strengthens the understanding  
of EV cybersecurity, it does not address diagnostic capabilities directly  
XII. Zhang et al. (2023) provided a broad review of AI integration in EV technologies [12]. Their work  
highlights the role of AI in EV development but lacks depth in machine learning applications for predictive  
monitoring.  
XIII.  
Das et al. (2023) applied DBN and LSTM models for accurate RUL estimation of EV batteries [13].  
Although effective, the implementation faces limitations due to edge device constraints.  
XIV. Wiki Contributors (2023) proposed a conceptual framework for integrated vehicle health management  
[14]. The framework outlines a system-wide monitoring structure but does not include practical hardware  
considerations.  
XV. Another contribution by Wiki Contributors (2023) outlined strategies for Intelligent Maintenance Systems  
(IMS) [15]. While useful for designing high- level predictive systems, the study is not AI-specific.  
XVI.  
Further, Wiki Contributors (2023) discussed Industrial Internet of Things (IIoT) architectures for  
predictive maintenance [16]. Though not designed specifically for EVs, these architectures can facilitate sensor  
integration in vehicular networks.  
XVII.  
Zhao et al. (2020) described an edge AI framework for vehicular applications [17]. Their work  
demonstrates scalable deployment of AI at the edge, though it is not directly tailored to EVs.  
XVIII.  
Mishra and Rao (2020) explored distributed vehicular computing through fog and AI technologies [18].  
Their approach supports distributed EV monitoring but suffers from scalability challenges.  
XIX.  
Chen et al. (2020) analyzed vulnerabilities in CAN bus systems and proposed IDS-based countermeasures  
[19]. While the study is vital for vehicle security, it does not integrate predictive maintenance aspects.  
XX.  
Finally, Yadav et al. (2020) investigated machine learning-based IDS techniques for real-time fault  
detection in CAN networks [20]. Although effective in detecting attacks, the work does not directly address  
health diagnostics.  
Summary:  
The existing studies on Predictive Health Monitoring (PHM) in Electric Vehicles (EVs) demonstrate the use  
of a variety of modern techniques designed to enhance vehicle reliability, safety, and performance. Researchers  
have implemented multiple approaches such as deep learning, anomaly detection, reinforcement learning,  
and digital twin systems to improve fault prediction and maintenance efficiency.  
Deep learning-based models process large amounts of vehicle data—particularly CAN bus signals—to  
estimate parameters like the State of Health (SoH) or Remaining Useful Life (RUL) of critical components  
including batteries, motors, and inverters. These models can automatically detect patterns in data and offer  
accurate health predictions under varying driving conditions.  
Anomaly detection methods are applied to identify abnormal behaviours or irregular sensor readings in real-  
time. Early identification of such deviations enables preventive maintenance, reduces breakdown risks, and  
ensures vehicle safety.  
www.ijltemas.in  
Page 1536  
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,  
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)  
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XIV, Issue XII, December 2025  
Reinforcement learning allows systems to continuously learn from operational feedback and make data-driven  
decisions to improve energy efficiency, fault recovery, and system stability. This adaptive nature helps maintain  
optimal performance over time.  
Digital twin technology involves building a virtual model of vehicle components that mirrors real-time  
operations. This virtual replica supports continuous tracking, simulation, and prediction of component wear and  
degradation, leading to better diagnostic and maintenance planning.  
However, despite these technological advancements, several limitations still remain. One major issue is the lack  
of standardized and open-access datasets, which restricts fair comparison and validation of different PHM  
techniques. Additionally, most existing systems are designed for cloud platforms rather than edge computing,  
which limits their ability to operate directly on low-power embedded devices inside the vehicle.  
Another key challenge is the insufficient real-world testing of proposed methods. Many models are validated  
only in controlled or simulated environments, which may not fully capture the complexities of real driving  
conditions.  
In conclusion, although notable progress has been made in predictive health monitoring for electric vehicles, the  
next step is to develop scalable, reliable, and secure PHM frameworks that are practical for deployment in real-  
world automotive systems.  
Author(s), Year  
Method / Technique  
Key Contribution  
ation / Future Scope  
I. Patil et al., 2025  
LSTM, Federated Learning Secure EV fault prediction High computational  
demand; needs  
optimization  
II. Wang et al., 2025  
SVM, K-Means  
OBD-II based emission &Requires standardized  
behaviour diagnostics  
OBD- II data  
III. Ahmed & Patel, 2025 Quantum AI +  
Federated Learning  
High accuracy in EV  
fault detection  
Relies on experimental  
quantum hardware  
IV. Roy et al., 2025  
V. Zhang et al., 2024  
VI. Lee et al., 2024  
VII. Singh et al., 2024  
Homomorphic Encryption Privacy-preserving CAN  
+ Edge AI diagnostics  
High processing load on  
devices  
Deep Learning + CAN Bus Survey  
detection  
on  
anomaly Lacks real-world testing  
DBN, RNN  
Accurate battery State of Not optimized for  
Health (SoH) prediction  
embedded devices  
Reinforcement Learning  
Hybrid vehicle energy  
optimization  
Not directly health-focused  
VIII. Fernandez et al., 2024Digital Twin Framework  
IX. Gupta & Sharma, 2023 Statistical + ML Models  
Virtual testing of EV  
powertrains  
Complex to implement  
RUL estimation from  
sensor data  
Limited  
real-time  
applicability  
X. Hossain et al., 2023  
CNN, SVM, Time Series  
Industrial  
predictive Not specific to EV systems  
maintenance methods  
www.ijltemas.in  
Page 1537  
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,  
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)  
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XIV, Issue XII, December 2025  
XI. Kumar & Singh, 2023 Security Protocols, IDS  
Review of CAN bus  
cybersecurity  
Does not address health  
monitoring  
XII. Zhang et al., 2023  
XIII. Das et al., 2023  
AI Technology Review  
DBN + LSTM  
Overview of AI in EVs  
Lacks ML-specific detail  
Accurate EV battery RULLimited by edge  
prediction  
hardware constraints  
XIV. Wiki Contributors, Conceptual Framework  
2023  
Integrated vehicle health  
monitoring  
No  
hardware  
implementation  
XV. Wiki Contributors,  
2023  
Predictive Strategies  
Intelligent  
system (IMS)  
maintenance General strategies, not  
AI- specific  
XVI. Wiki Contributors, IIoT Architecture  
2023  
Predictive  
framework  
maintenance Not EV-specific  
XVII. Zhao et al., 2022  
Edge AI Framework  
Vehicular edge  
computing for AI  
Not tailored to EVs  
XVIII. Mishra & Rao,  
2022  
Distributed AI + Fog  
Computing  
Distributed EV monitoring Scalability challenges  
XIX. Chen et al., 2022  
IDS-based Security  
Analysis  
of  
CAN  
No  
predictive health  
vulnerabilities  
monitoring integration  
XX. Yadav et al., 2022  
ML-based IDS  
Real-time CAN anomalyNot combined with health  
detection diagnostics  
Taxonomy of Approaches  
From the surveyed literature, PHM approaches can be broadly classified into:  
1. Model-Based Approaches: Simulation environments and digital twins for predicting potential failures.  
2. Data-Driven Approaches: ML and DL techniques such as LSTM, CNN, and DBN applied to CAN bus  
data.  
3. Security-Focused Approaches: IDS and encryption-based frameworks for secure data transfer.  
4. Edge-Based Implementations: Deployment of lightweight ML models on embedded devices for real-  
time fault detection.  
Comparative Analysis  
Focus Areas: Battery health monitoring and anomaly detection dominate research efforts.  
Trends: Research peaked in 2023, with sustained developments in 2024 and 2025.  
Strengths: Accuracy improvements through ML/DL, reinforcement learning in hybrid systems, and  
integration of digital twins.  
Weaknesses: Limited real-world deployment, lack of standardized datasets, and resource constraints  
on edge hardware.  
www.ijltemas.in  
Page 1538  
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,  
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)  
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XIV, Issue XII, December 2025  
Distribution of Research Focus in PHM of EV Powertrains (2020-2025)  
9
8
7
6
5
4
3
2
1
0
8
5
3
2
Motor &  
Powertrain  
Security &  
Privacy  
General  
Predictive  
Maintenance  
Battery  
Health  
Research Trends in Predictive Health Monitoring of Powertrains (2020-2025)  
8
7
6
5
4
3
2
1
0
2020  
2023  
20224  
2025  
Hardware Implementations  
Several studies have demonstrated the feasibility of deploying PHM systems on embedded hardware:  
Hardware Platforms: Raspberry Pi (5), MCP2515 CAN interface modules.  
Data Acquisition: CAN signals accessed via the OBD-II port for real-time monitoring.  
www.ijltemas.in  
Page 1539  
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,  
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)  
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XIV, Issue XII, December 2025  
Software Tools: Python with TensorFlow Lite, Scikit-learn, and python-can.  
Algorithms: One-Class SVM, Isolation Forest, and lightweight LSTM variants.  
Output Interfaces: Dashboards, LEDs, or mobile apps for driver alerts.  
Future Directions  
Standardized Datasets: Development of open CAN bus datasets for benchmarking predictive models.  
Optimized Edge Models: Model compression and pruning techniques for resource-constrained  
hardware.  
Digital Twins: Virtual replicas for safe and accurate fault simulation.  
Explainable AI: Enhancing interpretability and trust in fault predictions.  
Cybersecurity Integration: Strengthening PHM systems against in-vehicle cyber threats.  
Hybrid Edge–Cloud Frameworks: Combining edge responsiveness with cloud scalability.  
CONCLUSION  
This paper surveyed predictive health monitoring methods for EV powertrains using Edge AI and CAN bus data.  
The review highlighted progress in battery diagnostics, anomaly detection, security, and real-time  
implementations. Edge computing offers significant advantages over cloud-based systems, including reduced  
latency, privacy, and cost-effectiveness. However, gaps remain in dataset availability, scalability, and model  
optimization. Addressing these challenges will be critical to establishing PHM as a standard feature in future  
EV ecosystems.  
REFERENCES  
1. S. Patil, A. Mehta, and R. Kulkarni, “AI-driven predictive maintenance in electric vehicles using LSTM  
and federated learning,” in Proc. Int. Conf. Smart Mobility Syst., 2025, pp. 1–6.  
2. H. Wang, Y. Li, and P. Zhang, “OBD-II based machine learning applications for vehicle monitoring,”  
IEEE Sensors Journal, vol. 25, no. 4, pp. 1123–1132, 2025.  
3. M. Ahmed and D. Patel, “Quantum-enhanced artificial intelligence for fault diagnosis in electric  
vehicles,” IEEE Transactions on Vehicular Technology, vol. 74, no. 2, pp. 456–467, 2025.  
4. A. Roy, K. Sharma, and P. Nair, “Privacy-preserving predictive maintenance using homomorphic  
encryption and edge AI,” in Proc. IEEE Int. Conf. Edge Computing, 2025, pp. 201–208.  
5. Y. Zhang, T. Chen, and S. Huang, “A survey of anomaly detection in in-vehicle networks,” IEEE Access,  
vol. 12, pp. 54123–54139, 2024.  
6. J. Lee, R. Park, and H. Kim, “AI for battery health monitoring using deep belief networks and recurrent  
neural networks,” Energies, vol. 17, no. 6, pp. 3112– 3128, 2024.  
7. R. Singh, P. Verma, and S. Gupta, “Reinforcement learning for energy management in hybrid vehicles,”  
Chinese Journal of Mechanical Engineering, vol. 37, no. 2, pp. 215–225, 2024.  
8. M. Fernandez, L. Torres, and G. Romero, “Digital twin framework for EV powertrains,” World Electric  
Vehicle Journal, vol. 15, no. 5, pp. 234–245, 2024.  
9. P. Gupta and A. Sharma, “Predictive maintenance in the automotive sector using machine learning and  
statistical models,” Journal of Industrial Engineering Research, vol. 62, no. 3, pp. 154–166, 2023.  
10. T. Hossain, N. Alam, and M. Rahman, “AI applications in industrial equipment maintenance using CNN  
and SVM,” IEEE Transactions on Industrial Informatics, vol. 19, no. 7, pp. 8452– 8464, 2023.  
11. V. Kumar and S. Singh, “Cybersecurity in electric vehicles: A review of CAN bus protocols and intrusion  
detection systems,” IEEE Transactions on Intelligent Transportation Systems, vol. 24, no. 1, pp. 122–  
135, 2023.  
www.ijltemas.in  
Page 1540  
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,  
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)  
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XIV, Issue XII, December 2025  
12. H. Zhang, J. Liu, and F. Zhao, “AI integration in electric vehicle technology: A comprehensive review,”  
IEEE Access, vol. 11, pp. 78232–78249, 2023.  
13. S. Das, A. Mondal, and K. Banerjee, “Remaining useful life estimation using DBN and LSTM models,”  
in Proc. Int. Conf. Prognostics and Health Management (PHM), 2023, pp. 211–219.  
14. “Integrated vehicle health management: A conceptual framework,” Wikipedia, 2023. [Online].  
Available: https://en.wikipedia.org/wiki/Integrated_vehicle_hea lth_management. [Accessed: Oct. 18,  
2025].  
15. “Intelligent  
maintenance  
system  
(IMS),”  
Wikipedia,  
2023.  
[Online].  
Available:  
https://en.wikipedia.org/wiki/Intelligent_maintenanc e_system. [Accessed: Oct. 18, 2025].  
16. “Industrial Internet of Things (IIoT) for predictive maintenance,” Wikipedia, 2023. [Online]. Available:  
https://en.wikipedia.org/wiki/Industrial_Internet_of_ Things. [Accessed: Oct. 18, 2025].  
17. L. Zhao, Q. Wang, and J. Sun, “Vehicular edge computing and networking framework for AI  
deployment,” IEEE Internet of Things Journal, vol. 9, no. 8, pp. 7120–7134, 2022.  
18. P. Mishra and K. Rao, “Distributed vehicular computing for predictive maintenance using fog AI,” in  
Proc. IEEE Int. Conf. Distributed Systems, 2022, pp. 134–140.  
19. Y. Chen, L. Wu, and D. Xu, “CAN bus cyberattacks and countermeasures using intrusion detection  
systems,” in Proc. IEEE Vehicular Networking Conf., 2022, pp. 88–95.  
20. R. Yadav, M. Sharma, and K. Singh, “Machine learning-based intrusion detection techniques for real-  
time CAN bus security,” Journal of Embedded Systems and Vehicular Security, vol. 14, no. 2, pp. 99–  
110, 2022.  
www.ijltemas.in  
Page 1541