Work

Research on intelligent exploration robotic vehicle for underground utility tunnels

Year: 2026
Type: conference-paper
Abstract: With the rapid development of society, the construction of underground pipe galleries is becoming increasingly large, but traditional manual inspection has low efficiency, high cost, and high risk. Robotics research is a critical and rapidly developing field, and using robotic vehicles to survey underground pipe galleries is a viable approach. This study builds a digital twin platform and relies on multi-source data fusion to construct a three-dimensional visualization model of underground pipe galleries, with the synchronization latency between the robotic vehicle and the 3D visualization model controlled within 200 ms, achieving hazard prediction and maintenance decision support. The system adopts sound array imaging technology (FLYTEK AI 30), integrates multi-sensor modules, and builds fault classification and path optimization models based on deep learning frameworks (CNN+RNN). Anomaly recognition relies on the Recurrent Neural Network (RNN) model. The terminal monitoring platform is connected to the overall system of the pipe gallery center through IoT technology, and displays real-time abnormal alarm information. The experimental results show that the system significantly improves detection efficiency, reduces operating costs, and enhances security. This study effectively promotes the intelligent development of underground pipe gallery detection.
Language: English
Open Access status: hybrid
OpenAlex ID: W7204202138
FWCI:
Cites: 2
Cited by:
Related to: 10