Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

ETLNet: An Efficient TCN-BiLSTM Network for Road Anomaly Detection Using Smartphone Sensors

Mohd Faiz Ansari, Rakshit Sandilya, Mohammed Javed, David Doermann

TL;DR

ETLNet addresses road anomaly detection, specifically speed bumps, by leveraging smartphone inertial sensors to overcome lighting and markings limitations of vision-based systems. It fuses two Temporal Convolutional Network layers with a Bidirectional LSTM to extract temporal patterns from seven features collected at the dashboard-right location, evaluated on the Passive Vehicular Sensors Dataset. The model delivers near-perfect performance, with an F1-score of 99.3% for speed bumps and overall high accuracy (≈98.8%), outperforming prior methods across multiple sensor placements. This work enables scalable, crowdsourced road-condition monitoring via mobile devices, with potential for real-time alerts and cloud-based sharing to support safer, adaptive driving systems.

Abstract

Road anomalies can be defined as irregularities on the road surface or in the surface itself. Some may be intentional (such as speedbumps), accidental (such as materials falling off a truck), or the result of roads' excessive use or low or no maintenance, such as potholes. Despite their varying origins, these irregularities often harm vehicles substantially. Speed bumps are intentionally placed for safety but are dangerous due to their non-standard shape, size, and lack of proper markings. Potholes are unintentional and can also cause severe damage. To address the detection of these anomalies, we need an automated road monitoring system. Today, various systems exist that use visual information to track these anomalies. Still, due to poor lighting conditions and improper or missing markings, they may go undetected and have severe consequences for public transport, automated vehicles, etc. In this paper, the Enhanced Temporal-BiLSTM Network (ETLNet) is introduced as a novel approach that integrates two Temporal Convolutional Network (TCN) layers with a Bidirectional Long Short-Term Memory (BiLSTM) layer. This combination is tailored to detect anomalies effectively irrespective of lighting conditions, as it depends not on visuals but smartphone inertial sensor data. Our methodology employs accelerometer and gyroscope sensors, typically in smartphones, to gather data on road conditions. Empirical evaluations demonstrate that the ETLNet model maintains an F1-score for detecting speed bumps of 99.3%. The ETLNet model's robustness and efficiency significantly advance automated road surface monitoring technologies.

ETLNet: An Efficient TCN-BiLSTM Network for Road Anomaly Detection Using Smartphone Sensors

TL;DR

ETLNet addresses road anomaly detection, specifically speed bumps, by leveraging smartphone inertial sensors to overcome lighting and markings limitations of vision-based systems. It fuses two Temporal Convolutional Network layers with a Bidirectional LSTM to extract temporal patterns from seven features collected at the dashboard-right location, evaluated on the Passive Vehicular Sensors Dataset. The model delivers near-perfect performance, with an F1-score of 99.3% for speed bumps and overall high accuracy (≈98.8%), outperforming prior methods across multiple sensor placements. This work enables scalable, crowdsourced road-condition monitoring via mobile devices, with potential for real-time alerts and cloud-based sharing to support safer, adaptive driving systems.

Abstract

Road anomalies can be defined as irregularities on the road surface or in the surface itself. Some may be intentional (such as speedbumps), accidental (such as materials falling off a truck), or the result of roads' excessive use or low or no maintenance, such as potholes. Despite their varying origins, these irregularities often harm vehicles substantially. Speed bumps are intentionally placed for safety but are dangerous due to their non-standard shape, size, and lack of proper markings. Potholes are unintentional and can also cause severe damage. To address the detection of these anomalies, we need an automated road monitoring system. Today, various systems exist that use visual information to track these anomalies. Still, due to poor lighting conditions and improper or missing markings, they may go undetected and have severe consequences for public transport, automated vehicles, etc. In this paper, the Enhanced Temporal-BiLSTM Network (ETLNet) is introduced as a novel approach that integrates two Temporal Convolutional Network (TCN) layers with a Bidirectional Long Short-Term Memory (BiLSTM) layer. This combination is tailored to detect anomalies effectively irrespective of lighting conditions, as it depends not on visuals but smartphone inertial sensor data. Our methodology employs accelerometer and gyroscope sensors, typically in smartphones, to gather data on road conditions. Empirical evaluations demonstrate that the ETLNet model maintains an F1-score for detecting speed bumps of 99.3%. The ETLNet model's robustness and efficiency significantly advance automated road surface monitoring technologies.

Paper Structure

This paper contains 16 sections, 4 figures, 4 tables.

Table of Contents

  1. Introduction
  2. Related Work
  3. Proposed Method
  4. Experimental Protocol
  5. Dataset
  6. Training and Testing
  7. Evaluation Metrics
  8. Experiments
  9. Comprehensive Analysis of Model Efficacy Across Varied Window Sizes and Sensor Locations
  10. Experiment 1:
  11. Experiment 2:
  12. Experiment 3:
  13. Comparative Analysis of Speed Bump Detection Models
  14. Discussion and Conclusion
  15. Conclusion
  16. ...and 1 more sections

Figures (4)

  • Figure 1: Flowchart illustrating the process of speed bump detection.
  • Figure 2: Proposed architecture of the ETLNet model for speed bump detection.
  • Figure 3: Model Performance
  • Figure 4: Line graphs showing the validation accuracy and F1-scores across different window sizes for each model.