Wireless Intelligent Accurate Bridge Deck Crack Inspection and Mapping

Abstract

One of the important tasks for building maintenance is building deck crack inspection. However, the inspection result in human analysis is less accurate in nature. So, a crack inspection system that uses a camera-equipped mobile robot to collect images on the building deck has been proposed. In this method, the Laplacian of Gaussian (LoG) algorithm is used to detect cracks and a global crack map is obtained through camera calibration and robot localization. To clarify that, the robot collects all the images on the building deck, and a path planning algorithm based on the genetic algorithm is developed. We validate our proposed system through both, simulations and experiments. This work addresses crack detection and mapping on a building deck using a robotic system. Several challenges including coordinate transformation, robot localization and complete coverage path planning for the proposed robot system are tackled. This paper focuses mainly on the overall framework for such a robotic inspection system; therefore some of the techniques for handling shadows, paints, patches on buildings are not addressed. In real-world applications, these issues should be carefully incorporated into the design of the image processing algorithm. Also, there may be vibration caused by the passing traffic, which should be dealt with as well. The positioning of the ROCIM system is critical to crack mapping, hence more accurate robot localization techniques fusing various sensors such as differential GPS, inertial measurement unit (IMU), etc. should be developed. It is also worth noting that the depth and severity of the cracks can be measured by employing advanced nondestructive evaluation (NDE) sensors, such as impact echo and ultrasonic surface wave.

Keywords: Robots, mapping systems, ROCIM system

Cite this Article
Mani Chandana R, Srinivas P. Wireless Intelligent Accurate Bridge Deck Crack Inspection and Mapping. Journal of Advancements in Robotics. 2016; 3(2): 24–29p.