Development of an Automated Laser Depaint System, 10-R8155

Printer Friendly Version

Principal Investigators
Dan H. Weissling
Michael P. Rigney
Michael O. Blanton Jr.

Inclusive Dates:  06/03/10 – Current

Background - Laser paint removal has the potential to become the process of choice for a number of industries. The advantages of laser coating removal include reduced waste stream volume, lower maintenance and lower power consumption. Although these advantages were recognized two decades ago, the ability to deliver sufficient laser power to the end effector of an extended-reach robotic system suitable for whole-aircraft paint removal has only recently become feasible with the maturity of fiber lasers. In addition, this type of laser offers power control capability appropriate to prevent damage to sensitive substrates and enable selective coating removal. Though manual open-loop testing is showing good results, the key to the success of an automated system is closed-loop sensing/control of the process. The purpose of this project is to develop a control solution that comprises sensors and data analysis software to measure the coating removal state and provide real-time feedback to modulate laser power. This closed-loop control system will then be integrated with a commercial fiber laser into a custom robotic platform.

Approach - This project will develop a real-time, closed-loop control system that will use machine vision to characterize the target surface, and then provide a control signal to modulate a laser power source. A high frame rate color camera and a filtered monochrome camera are used to provide RGB and NIR spectral data for real-time analysis of the coating removal process. The two cameras are optically aligned via a beam splitter, and software controls allow for fine-tuning to match the field of view of both cameras. Images are acquired at twice the rate of the laser raster period. This allows alternate analysis of the left- and right-image fields, which are synchronized with the laser position. Subsequent analysis and spectral classification are used to identify coating vs. substrate and modulate the laser power to precisely control the coating removal process. Work will be executed in two phases: Phase 1 will develop and validate the sensor/control system, and Phase 2 will integrate the new technology into a large-scale coating-removal robotic platform at SwRI to validate and demonstrate system capability.

Accomplishments - To date, the design and assembly of the test bed were completed. Laboratory testing began in October 2010.

2010 Program Home