Recent developments in robotic systems, automation and computer vision and sensors have well prepared the ground towards automated robotic solutions for inspection of civil infrastructures and particularly tunnels ageing urgently requiring serious inspection and assessment. Nowadays, tunnel inspections are being performed manually and visually by human operators.
ROBO-SPECT is an EC co-funded research project (FP7 - ICT – 611145) that is driven by the tunnel inspection industry and that adapts and integrates recent research results in intelligent control in robotics, computer vision and active continuous learning and sensing, in an innovative, integrated, robotic system. ROBO-SPECT automatically scans the internal surface of tunnels for potential defects (spalling, delamination etc) and inspects and measures radial deformation in the cross-section, distance between parallel cracks, cracks and open joints that impact tunnel stability, with mm accuracies.
The robotic system (currently at a prototyping phase) consists of an intelligent robotic system supported by advanced control systems and an on-board high precision robotic tip, an advanced computer vision system based on high quality and stereo cameras for cracks/defects detection and 3D vision, an ultrasonic sensor able to detect crack width and depth with high accuracy. The system also includes a 3D laser scanner able to perform tunnel deformation measurements also with high accuracy. This publication describes the integration status and recent developments of the semi-autonomous robotic system as designed for concrete lining tunnels. The final system architecture, communication mechanisms and integration steps of the whole robotic system are also being presented.