Path-Planning Algorithm for Automated Pavement Crack Sealing Based on Postman Problems

Abstract

Cracks exist in many civil infrastructures, such as road and bridge deck surfaces, parking lots, and building surfaces. To prevent crack growth and further deterioration, it is necessary to fill these cracks with appropriate materials in a timely manner. To improve maintenance efficiency, many machines that automatically complete sealing work have been developed in construction and maintenance areas. However, most current path-planning algorithms used in machines are based on transverse and longitudinal cracks with relatively simple shape. This article presents a new path-planning algorithm for complex cracks that is used for automatic pavement crack-sealing systems. Information about the crack binary map is extracted from the actual crack image using our previous research work as a guide. Crack information is extracted and complex cracks are classified based on graph theory. The path-planning problem of connected cracks is then simplified to the Chinese Postman Problem, and the path-planning problem of unconnected cracks is simplified to the Rural Postman Problem. Finally, the feasibility of using the algorithm to deal with complex cracks is verified at the software level. The results indicate that the proposed algorithm has lower complexity and greater efficiency.