Reinforced concrete is one of the most commonly used materials in construction for example buildings, bridges, sidewalks, power plants, and dams. These structures are challenged particularly with surface cracking resulting in common life-cycle-related damage that may jeopardize their serviceability and safety. It is difficult, time-consuming, and sometimes impossible to identify small cracks by the naked eye. Besides, cracks in structures like dams and power plants are harder to access compared to sidewalks and buildings. Cracks in concrete structures can be indicators of crucial damage and may negatively affect durability. To ensure the overall soundness of these structures surface cracks have to be detected promptly. In recent years, computer vision algorithms have made automatic detection of such cracks possible. However, the current practice for repairing them is manual and labor intensive.To automate and integrate the process of crack detection and repair, this paper presents a robot with the ability to detect and repair surface cracks. The robot was designed, manufactured, and tested as part of a senior design project for undergraduate students. To detect the crack, this robot utilizes an RGB camera, a stereo infrared depth sensor, and the LIDAR technology. To fuse the multi-sensor data, this study uses a Bayesian network. In particular, the shape, hue, and depth of cracks are used to probabilistically detect crack-related pixels. The robot then uses the detection results to automatically fill the cracks with commercial concrete crack filler materials. In particular, the robot is equipped with a costume-designed piston and syringe mechanism that injects the filler material in a controlled manner. To push the piston this robot uses a linear actuator mechanism that is consists of a non-captive lead screw and a stepper motor. During the injection, the robot positions itself on the start of the crack and adjusts its alignment in such a way that its tank-based wheels are parallel to the crack. In this way, it follows the unfilled portion of the crack. To accurately position the injector along the crack the robot utilizes a two DOF arm that allows rotation with respect to the vertical axis of the robot and extension along the arm. To control the arm, the same data-fusion algorithm and vision-system used for crack detection are used as a feedback loop. This ensures the uniformity and consistency of the repaired concrete surface.To evaluate the performance of the robot, a simulated cracked surface was carved out of Styrofoam and placed on a plywood platform. Specifically, the simulated crack included a continuous crack whose width and depth ranged from 3 mm to 6 mm at different locations. The results show the robot can accurately detect cracks that their width is in the millimeter range. It can also fill the cracks without under-filling or overfilling. In this way, the robot successfully automated the currently manual repair process for surface concrete cracks. Despite the encouraging results and their potential impact on the industry, this study had some limitations, in particular, the performance of the robot was only tested on a single and simulated crack. In the future, the effectiveness of the robot should be verified on real concrete surfaces.

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

Paper Proceedings: https://asmedigitalcollection.asme.org/IMECE/IMECE2021/volume/85543

ASME 2021 International Mechanical Engineering Congress and Exposition (IMECE2021)

Supporting the need among researchers worldwide for additional online and multimedia content, ASME now offers the ASME Conference Video Collection. This platform will collect and provide long-term access to conference presentation videos, encompassing all technology areas within ASME's technical conference and events program. About 30,000-40,000 engineers across all mechanical engineering and related disciplines participate in ASME conferences each year. This includes authors of approximately 6,000 to 8,000 technical papers published by ASME, the majority of whom also submit video presentations for our virtual conferences. The video collection will be available as an add-on to existing product packages or as a separate subscription.