Abstract
A three-dimensional measuring method for pavement based on multi laser displacement sensors is proposed to establish a complete and powerful database of pavement shape. Model of the measuring system based on this method is established and analyzed. Data generated by sensors, including data of laser displacement sensors, acceleration sensors and gyroscope, is transmitted according to TCP/IP Protocol. The back-end computer fuses the data and establishes the databank of 3-D pavement shape. Experimental results of pavement shape measurement using gradienter and the system based on this method proved that the application and accuracy of this method are satisfied.
Introduction
Maintaining safe and comfortable road infrastructure requires continuous monitoring of pavement quality. Traditionally, engineers rely on two primary metrics: the International Roughness Index (IRI) for longitudinal "bumpiness" and rut depth for transverse depressions caused by tire wear. However, existing tools often provide only 2D snapshots—either a single longitudinal line or isolated cross-sections. To achieve a comprehensive understanding of road health, a full 3D database of pavement shape is required.
The primary challenge in high-speed 3D mapping is vehicle dynamics. As a measuring vehicle moves, it vibrates, tilts, and bounces. These movements introduce significant noise into laser displacement data, making it difficult to distinguish between actual road distress and vehicle-induced motion.
System Design and Hardware
The proposed system utilizes a mobile platform equipped with a rigid, extendable measuring beam. To cover a full 3.75-meter lane width while remaining safe for transport, the researchers developed a two-layer beam design consisting of an "uplayer" and "sublayer" that can retract when not in use.
The sensor suite includes:
- Multi-Laser Displacement Sensors (LDSs): Distributed non-equidistantly, with higher density over expected rutting areas to optimize data capture.
- Acceleration Sensors: Mounted at the beam's extremities to track high-frequency vibrations.
- Gyroscopes: Used to determine the overall orientation and geographic trend of the vehicle.
3D Modeling and Data Fusion
The core innovation lies in the mathematical correction of the "floating benchmark." Because the measuring beam is not perfectly stable, the system treats its movement as a variable. By fusing high-frequency accelerometer data with laser readings, the system calculates the beam's deviation from an ideal benchmark.
Using the Least Squares Method, the system filters out vehicle motion to generate a continuous 3D surface $R$. This allows for the extraction of both longitudinal profiles and cross-sectional curves at any point along the surveyed route, effectively bridging the gap between isolated 2D measurements and a holistic 3D model.
Experimental Validation
The system was validated against a ground-truth benchmark created by manually measuring six 30-meter road sections using precision gradienters on a dense 0.25m x 0.25m grid. Real-world testing was conducted on several Chinese expressways, including the Anyang-Hebi and Weiwu-Gulang routes.
At a testing speed of 72 km/h and a sampling interval of 0.05m, the system demonstrated high precision:
- IRI Relative Error: Less than 3%.
- Rut Depth Relative Error: Less than 2%.
Conclusion
The multi-laser displacement sensor system provides a robust solution for high-speed 3D pavement mapping. By successfully correcting for vehicle dynamics through sensor fusion, it enables the creation of precise infrastructure databases. Future work will focus on the system's performance under varying environmental conditions and the integration of GPS/GNSS data for comprehensive geographic information system (GIS) mapping of national highway networks.
Summary updated