Optical simulation of scratch repair in F/A-18 transparences

icaf2023 Tracking Number 135

icaf2023 presentation

An on-aircraft system for assessing scratches and dents in F/A-18 transparencies (windshield and canopy) and applying optical simulation for optimizing their repair is presented. The motivation was to automatize the detection and quantification on transparency defects and to assist the manual repair process to minimize induced optical distortions by replacing visual inspection and subjective decision making with automatic, repeatable and objective solution. The information can be applied to optimize the whole logistics chain of the repair process and to maximize aircraft availability. The system developed consist of three consecutive measurement tasks: First a machine vision-based scratch/dent detection and mapping enabling detecting and classifying defects for further inspection. Second, the dimension of the chosen defects are measured. Third , based on the defect location and dimension, optical simulation is applied to estimate the effect of the repair process, and a most suitable one is chosen regarding optical distortion. The above on-aircraft machine vision system developed for defect detection automatically scans and subsequently maps all defects on the transparencies and provides estimates of their scattering from the pilot perspective. The resulting map enables automatic classification of the scratches and dents found, based on user requirements. It also enables monitoring of the transparency defects’ evolution through time, thus providing a modern tool for the transparencies’ life cycle management. In addition to the above, a manually operated optical micrometre was developed for measuring the depth of the scratch or dent, based on 3D imaging. Last but not least, optical simulation models were developed for connecting the optical distortion induced by the material extraction in the repair process and vice versa. With the simulation, the effect of the repair can be mapped into standardized optical distortion measurement for evaluation. The performance of the defect mapping and depth estimation was assessed by real transparent samples and reference measurements. The simulation models were validated by comparing them with a real defect repair with before and after measurement of optical distortion. The developed system will be fielded to operational use within the Finnish Air Force.