Perception and Control for Automated Pavement Crack Sealing
by Chris Hendrickson, Carnegie Mellon Univ, Pittsburgh, United States,Sue McNeil, Carnegie Mellon Univ, Pittsburgh, United States,
Darcy Bullock, Carnegie Mellon Univ, Pittsburgh, United States,
Carl Haas, Carnegie Mellon Univ, Pittsburgh, United States,
Daniel Peters, Carnegie Mellon Univ, Pittsburgh, United States,
Darrin Grove, Carnegie Mellon Univ, Pittsburgh, United States,
Kent Kenneally, Carnegie Mellon Univ, Pittsburgh, United States,
Shannon Wichman, Carnegie Mellon Univ, Pittsburgh, United States,
Document Type: Proceeding Paper
Part of: Applications of Advanced Technologies in Transportation Engineering
Abstract:
Pavement maintenance is a major transportation engineering activity that has seen only partial mechanization and virtually no automation. Crack sealing is a typical maintenance task in this area that has not been automated to any degree. In this paper, we report on a program of laboratory experiments to develop perception and control methods to enable effective automation of routed pavement crack sealing. These experiments included: Investigation of sensor options and combinations. Implementation, calibration and testing of various image processing steps. Registration and alignment of video and range sensing data to permit pavement crack modelling. Implementation and testing of real time effector controls for tracing pavement cracks. The series of experimental implementations led to the development of a laboratory prototype system for automated pavement crack sealing. Implications of our experiments for perception of general pavement distress and for control of maintenance equipment will be drawn with respect to automation feasibility, reliability of automated equipment, and the economic efficiency of such systems.
Subject Headings: Automation and robotics | Pavement condition | Cracking | Sealing | Maintenance and operation | Equipment and machinery | Computer vision and image processing
Services: Buy this book/Buy this article
Return to search