Repairing potholes is a task for municipalities to prevent serious road user injuries and vehicle damage. This study presents a low-cost, high-performance pothole monitoring system to maintain urban roads. The authors developed a methodology based on photogrammetry techniques to predict the pothole's shape and volume. A collection of overlapping 2D images shot by a Raspberry Pi Camera Module 3 connected to a Raspberry Pi 4 Model B has been used to create a pothole 3D model. The Raspberry-based configuration has been mounted on an autonomous and remote-controlled robot (developed in the InfraROB European project) to reduce workers' exposure to live traffic in survey activities and automate the process. The outputs of photogrammetry processing software have been validated through laboratory tests set as ground truth; the trial has been conducted on a tile made of asphalt mixture, reproducing a real pothole. Global Positioning System (GPS) and Geographical Information System (GIS) technologies allowed visualising potholes on a map with information about their centre, volume, backfill material, and an associated image. Ten on-site tests validated that the system works in an uncontrolled environment and not only in the laboratory. The results showed that the system is a valuable tool for monitoring road potholes taking into account construction workers' and road users' health and safety.
A Robotized Raspberry-Based System for Pothole 3D Reconstruction and Mapping / Bruno, Salvatore; Loprencipe, Giuseppe; DI MASCIO, Paola; Cantisani, Giuseppe; Fiore, Nicola; Polidori, Carlo; D’Andrea, Antonio; Moretti, Laura. - In: SENSORS. - ISSN 1424-8220. - 23:13(2023). [10.3390/s23135860]
A Robotized Raspberry-Based System for Pothole 3D Reconstruction and Mapping
Salvatore Bruno;Giuseppe Loprencipe;Paola Di Mascio;Giuseppe Cantisani;Nicola Fiore;Carlo Polidori;Antonio D’Andrea;Laura Moretti
2023
Abstract
Repairing potholes is a task for municipalities to prevent serious road user injuries and vehicle damage. This study presents a low-cost, high-performance pothole monitoring system to maintain urban roads. The authors developed a methodology based on photogrammetry techniques to predict the pothole's shape and volume. A collection of overlapping 2D images shot by a Raspberry Pi Camera Module 3 connected to a Raspberry Pi 4 Model B has been used to create a pothole 3D model. The Raspberry-based configuration has been mounted on an autonomous and remote-controlled robot (developed in the InfraROB European project) to reduce workers' exposure to live traffic in survey activities and automate the process. The outputs of photogrammetry processing software have been validated through laboratory tests set as ground truth; the trial has been conducted on a tile made of asphalt mixture, reproducing a real pothole. Global Positioning System (GPS) and Geographical Information System (GIS) technologies allowed visualising potholes on a map with information about their centre, volume, backfill material, and an associated image. Ten on-site tests validated that the system works in an uncontrolled environment and not only in the laboratory. The results showed that the system is a valuable tool for monitoring road potholes taking into account construction workers' and road users' health and safety.File | Dimensione | Formato | |
---|---|---|---|
Bruno_A-robotized-raspberry-based-system_2023.pdf
accesso aperto
Note: articolo
Tipologia:
Documento in Post-print (versione successiva alla peer review e accettata per la pubblicazione)
Licenza:
Creative commons
Dimensione
5.02 MB
Formato
Adobe PDF
|
5.02 MB | Adobe PDF |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.