Robotic-Based Repair of Concrete Structures: A Surface Crack Filler Robot

Abstract

Surface cracks in concrete structures are often indicators of more substantial damage and may negatively affect the durability of a structure. To ensure the soundness of these structures, surface cracks should be quickly detected; this project proposes a robot with the ability to detect, map, and fill surface cracks. The robot will use a Bayesian network to fuse the multi-sensor data provided via an RGB camera, a stereo infrared depth sensor, and a LIDAR sensor. It will also be fitted with a newly designed piston-driven syringe system to inject a concrete filler material in a controlled manner. A non-captive lead screw and stepper motor drive the piston along with the syringe, and an arm with two degrees of freedom will allow the robot to position the injector along a crack accurately. To control the arm, the Bayesian network and sensor systems will work in unison to determine when a crack has been filled in a satisfying manner, ensuring a degree of uniformity and consistency in the repaired concrete surface.