CORTEXA
← Browse
crossrefMachines2025-09-28Cited by 4

From Sensors to Insights: Interpretable Audio-Based Machine Learning for Real-Time Vehicle Fault and Emergency Sound Classification

Mahmoud Badawy, Amr Rashed, Amna Bamaqa, Hanaa A. Sayed, Rasha Elagamy, Malik Almaliki, Tamer Ahmed Farrag, Mostafa A. Elhosseini

Unrecognized mechanical faults and emergency sounds in vehicles can compromise safety, particularly for individuals with hearing impairments and in sound-insulated or autonomous driving environments. As intelligent transportation systems (ITSs) evolve, there is a growing need for inclusive, non-intrusive, and real-time diagnostic solutions that enhance situational awareness and accessibility. This study introduces an interpretable, sound-based machine learning framework to detect vehicle faults and emergency sound events using acoustic signals as a scalable diagnostic source. Three purpose-built datasets were developed: one for vehicular fault detection, another for emergency and environmental sounds, and a third integrating both to reflect real-world ITS acoustic scenarios. Audio data were preprocessed through normalization, resampling, and segmentation and transformed into numerical vectors using Mel-Frequency Cepstral Coefficients (MFCCs), Mel spectrograms, and Chroma features. To ensure performance and interpretability, feature selection was conducted using SHAP (explainability), Boruta (relevance), and ANOVA (statistical significance). A two-phase experimental workflow was implemented: Phase 1 evaluated 15 classical models, identifying ensemble classifiers and multi-layer perceptrons (MLPs) as top performers; Phase 2 applied advanced feature selection to refine model accuracy and transparency. Ensemble models such as Extra Trees, LightGBM, and XGBoost achieved over 91% accuracy and AUC scores exceeding 0.99. SHAP provided model transparency without performance loss, while ANOVA achieved high accuracy with fewer features. The proposed framework enhances accessibility by translating auditory alarms into visual/haptic alerts for hearing-impaired drivers and can be integrated into smart city ITS platforms via roadside monitoring systems.

View free PDFSource page

Related papers

openalexMachines2026-07-24

Adaptive Weight Generation Neural Network LQR Control for Energy-Regenerative Suspension

Buyun Zhang, Beibei Xu, Sunfeng Qian, Yunshun Zhang, Chin An Tan

Vehicle energy-regenerative suspension can convert part of the vibration energy induced by road excitation into electrical energy. However, there are coupled performance conflicts among energy recovery, ride comfort, and suspension safety, and a fixed-weight LQR controller finds…

View free PDFSource page
openalexMachines2026-07-23

A Hierarchical Shared Steering Control Strategy Based on Driver States

Quanjin Wang, Lina Xuan, Jiwei Feng, Jian Wu

Continuous shared control provides an effective approach for intelligent vehicles to balance driving autonomy and system safety boundaries in complex human–machine interaction scenarios. However, existing shared control methods fail to dynamically adapt to the complex and time-va…

View free PDFSource page
crossrefMachines2026-04-06

Hybrid-Mechanism Deep Learning Modeling for Machine Tool Thermal Error: Robust Prediction via Few-Sample Learning

Hongru Chen, Yubin Huang, Chaochao Qiu, Xueyan Ning, Pingjiang Wang, Ke Yang

To address spindle thermal error in precision machining, this study proposes a hybrid modeling method. It combines a physical model for linear deformation with a GAT-LSTM network. Experiments show the hybrid model achieved RMSE/MAE of 4.6/4.0 µm under full training (12 conditions…

View free PDFSource page
crossrefMachines2026-03-19Cited by 1

Developing a Digital Twin for Human Performance Assessment in Human–Machine Interaction

Erik Novak, Aljaž Javernik, Iztok Palčič, Robert Ojsteršek

Digital twins are becoming essential tools in smart, human-centric manufacturing, yet validated approaches that integrate real human behavior into digital twin models remain limited. This study develops and experimentally validates a digital twin as a tool for evaluating human pe…

View free PDFSource page
crossrefMachines2026-02-12Cited by 2

Interpretable and Noise-Robust Bearing Fault Diagnosis for CNC Machine Tools via Adaptive Shapelet-Based Deep Learning Model

Weiqi Hu, Huicheng Zhou, Jianzhong Yang

Rolling bearings are crucial components in CNC machine tool spindles, and their health condition directly affects machining precision and operational reliability. To address the significant challenges of bearing fault diagnosis in industrial environments, this paper proposes an a…

View free PDFSource page
crossrefMachines2026-01-07Cited by 2

Cooperative Control and Energy Management for Autonomous Hybrid Electric Vehicles Using Machine Learning

Jewaliddin Shaik, Sri Phani Krishna Karri, Anugula Rajamallaiah, Kishore Bingi, Ramani Kannan

The growing deployment of connected and autonomous vehicles (CAVs) requires coordinated control strategies that jointly address safety, mobility, and energy efficiency. This paper presents a novel two-stage cooperative control framework for autonomous hybrid electric vehicle (HEV…

View free PDFSource page