Recent camera product introductions are highlighting the importance of machine vision in aerospace manufacturing. At the Vision 2011 conference in Stuttgart, Germany, Microscan demonstrated its Vision MINI and Vision HAWK smart cameras.
The small size of both makes them useful for a wide range of identification and inspection applications, including the meticulous lot tracking and component traceability required in aircraft assembly. The MINI measures 1.80in x 2.10in x 1.00in, and the HAWK measures 1.59in x 2.27in x 3.79in. Other recent introductions include the Imperx 29-megapixel ICL-B6620, a very high-resolution camera measuring 2.36in x 2.36in x 1.77in.
Inspection in aerospace manufacturing sometimes requires very small, very high-resolution video cameras,
such as this 60 x 60 x 45mm, 29-megapixel B6620 camera from Imperx.
Photo courtesy of Imperx
High quality and precision are top requirements in the components and processes used in automated assembly lines for aerospace manufacturing. Aircraft manufacturers are required to mark each component that goes into an airplane assembly with a 2D code and track it throughout the process. Their manufacturing facilities need machine vision not only for inspection, but also for automated tracking, tracing, and control of a wide range of part and assembly sizes and shapes.
These separate but interrelated functions help ensure quality, precision, accuracy, and traceability. Individual components, component lots, and assemblies must be tracked and traced, often over long periods of time, and problems must be easily and quickly identified, located, and isolated in order to be controlled.
A complex production environment requires a system for tracking the real-time location of sub-components, components, and sub-assemblies at any given point in the manufacturing process. Next is a system for tracing where they were before that given time and what processes were done to them there. This can provide a lifecycle history of the items. The third element is a system for determining and controlling where parts will go next in the process, based on whether they meet the quality criteria to continue to the next step. All three of these systems can be merged into a single machine vision-assisted system, which can optionally perform inspections.
In aerospace, the steps in this overall vision-enabled tracking, tracing, and control system are sometimes identified as part identification or recognition, part acceptance or rejection, part request or handoff, deformation detection, and gap and flush measurement. For part identification, parameter tolerance data must be input, such as location, width, height, depth, and diameter. To accept or reject a part, the features of a scanned part must be compared to their predefined parameters. Components that pass the parameter test will be handed off to another station for assembly. To do this correctly, other features of the part -- usually CAD dimensions -- must be compared to those of the requesting station to ensure that it's the right one for the job.
In deformation detection, things such as impact damage or corrosion in panels, rotor blades, or other components are detected by measuring the planarity or curvature of a surface. Those measurements are compared to tolerances for the particular component, and the extent to which it is out of plane or out of curvature is noted. Finally, gap and flush measurement is used to make sure mating assemblies will connect properly, such as mating dashboards to window frames, seats to interior cockpit frames, or doors to door frames.