Prototyping Fully Autonomous Bridge Visual Inspection based on mobile robots with visual recognition capabilities
|
|
|||||||||||
Bibliographic Details
| Author(s): |
Yasutaka Narazaki
(ZJU-UIUC Institute, Zhejiang University, Haining, Zhejiang, China.)
Mingyu Shi (ZJU-UIUC Institute, Zhejiang University, Haining, Zhejiang, China.) Linlong Meng (ZJU-UIUC Institute, Zhejiang University, Haining, Zhejiang, China.) |
||||
|---|---|---|---|---|---|
| Medium: | conference paper | ||||
| Language(s): | English | ||||
| Conference: | IABSE Congress: Beyond Structural Engineering in a Changing World, San José, Cost Rica, 25-27 Seotember 2024 | ||||
| Published in: | IABSE Congress San José 2024 | ||||
|
|||||
| Page(s): | 969-976 | ||||
| Total no. of pages: | 8 | ||||
| DOI: | 10.2749/sanjose.2024.0969 | ||||
| Abstract: |
This research discusses the technical roadmap and prototype development for replacing the human inspectors for bridge visual inspection by autonomous robots. The envisioned mobile robotic system first navigates and collects high-quality image data of critical structural components based on its own visual recognition of the bridge inspection scenes. The collected data is further post-processed to provide information about structural conditions. Preliminary results in structural component recognition and navigation planning steps are presented. Then, ongoing work to integrate those subsystems into a prototype autonomous system in the laboratory environment is discussed. Finally, the challenges that need to be addressed to realize such a complex autonomous system that performs tasks in the field environment are discussed. This work will motivate further investigations to accelerate the ongoing transformation in autonomous structural inspection. |
||||
| Keywords: |
bridge Inspection visual inspection deep learning autonomous structural inspection visual recognition mobile robots navigation planning synthetic environments
|
||||