For details, please refer to the paper
In this project, we explored the potential of using vehicle-mounted smartphones to estimate the International Roughness Index (IRI) for monitoring pavement roughness. Monitoring pavement roughness is critical for minimising vehicle damages and ensuring road user safety. Conventional roughness measurement instruments are costly and limited in surveying frequency and spatial coverage. To overcome these limitations, vehicle-mounted smartphones have been adopted to measure pavement roughness based on the dynamic responses of traversing vehicles. Nonetheless, the accuracy and consistency of the current smartphone-based approaches are affected by practical factors including speed, vehicle type and mounting configuration. Existing research applied deep learning to mitigate the impact of varying practical factors, but most of them was based on simulation studies. This study introduces a method of estimating the International Roughness Index (IRI) using smartphone-collected real vehicle response.
Research framework
Ground-truth IRI distributino on the experiment routes
 |
 |
 |
|---|---|---|
| Main | Vehicle select | Mount select |
Figure: Data collection App user interface
By utilising deep learning algorithms, specifically a multi-layer perceptron (MLP) model, we account for practical factors that affect smartphone-based IRI estimation, such as vehicle type, mounting configuration, and driving speed. The model was trained using real vehicle response data collected from multiple vehicles under various driving conditions. The results achieved a Root Mean Square Error (RMSE) of 0.60 and an R-squared value of 0.79 when compared with ground-truth IRI measurements.
Estimated IRI vs Ground truth IRI datapoints and regression line