Road Surface Condition Evaluation Using Imaging, LiDAR, and Multi-Grade Navigation Systems

Area of Science:

• Civil Engineering
• Transportation Engineering
• Geospatial Science

Background:

• Effective road surface condition monitoring is vital for transportation safety and efficiency.
• Existing methods for pavement anomaly detection have limitations in accuracy, cost, or scalability.

Purpose of the Study:

• To compare the performance of imagery-, LiDAR-, and accelerometer-based approaches for pavement anomaly detection.
• To evaluate the scalability of accelerometer-based monitoring for network-level road assessment.
• To integrate anomaly detection results into an interactive visualization tool.

Main Methods:

• Comparative analysis of three sensing modalities (imagery, LiDAR, accelerometer) on a 5-mile urban road segment.
• Evaluation of detection accuracy, severity classification, and processing efficiency for each method.
• Scalability testing of the accelerometer-based approach across a 36-mile roadway network.

Main Results:

• Accelerometer-based methods achieved the highest F1-scores (up to 97.2%), surpassing LiDAR (93.0%) and imagery (88.5%).
• The accelerometer approach demonstrated strong scalability and high spatial agreement across a 36-mile network.
• An integrated web portal facilitated interactive validation of detected pavement defects.

Conclusions:

• Accelerometer-based monitoring presents a practical, scalable, and cost-effective solution for widespread pavement evaluation.
• LiDAR and imagery serve as valuable complementary tools for detailed verification and characterization of road surface anomalies.
• Consumer-grade accelerometer sensors are viable for network-level pavement condition monitoring.