Rabbits Detected in the Blink of a Mechanical Eye
Friday, August 7, 2009
By running rabbits through Ford's Campbellfield production plant, Swinburne University of Technology researchers are working with the car manufacturer to test the performance of the next generation of visual inspection systems.
It's not rabbits of the animal kind, however, that are running loose—in the automotive industry a 'rabbit' is jargon for fault.
To gauge the limits of visual inspection systems, rabbits, including hard-to-see faults such as missing or unscrewed bolts, were deliberately released into cars on Ford's production line. The results showed that these systems can pick-up on some, but not all, faults. Essentially, non-contact inspection systems do as their name implies. Using techniques such as vision, ultrasonics and light scattering, they inspect components on a production line in real-time.
The use of these inspection systems is growing rapidly in the global automotive, aerospace and agricultural industries, and is an area Swinburne University of Technology's Non-Contact Inspection Research Group has been involved with since 1991.
Although vision inspection systems are applicable in many environments, the variations encountered in manufacturing can limit their use.
Group leader, Professor Romesh Nagarajah, said the objective of the project—funded by the Australian Research Council (ARC)—is to extend the reliable application of non-contact inspection systems by testing for robustness and accuracy in a real-life manufacturing environment. "In other words, the system must be able to accommodate the differences encountered in a manufacturing environment, be that from vibrations, dust or oil, light changes, or from natural component variations," Professor Nagarajah said.
"That is why we tend to carry out most of our work on the actual shop floor."
Professor Nagarajah added that, for Australian manufacturers, this type of inspection system must also be flexible enough to cope with small production runs and complex parts.
The Automotive Industry is a fiercely competitive business environment and has become even more volatile in the wake of the Global Financial Crisis. Australian manufacturers operate in a climate where many of its overseas rivals have much lower operating and labour costs. Visual inspection systems have the potential to play a crucial role in the future strategies of Australian vehicle-makers as they can reduce costs and improve quality—two essential ingredients in remaining competitive in the international automotive market. Ford already uses various types of vision systems for inspections at its plants in other countries.
Professor Nagarajah said a vision system typically comprises cameras, lighting systems and software for image processing, and analysis and decision making—deciding whether a component is 'good' or 'bad'.
Off-the-shelf cameras and lighting systems were configured by the research team and software platforms were developed to create a prototype non-contact inspection system that uses vision. Using this set-up, researchers are exploring the system’s limits by inspecting nine under-body car components in real-time, representing varying degrees of complexity.
"If you are going to install these types of systems on a production line, they have to achieve the speeds currently achieved on the line... or faster," Professor Nagarajah said.
"In most cases automated systems replace humans, delivering both positives and negatives.
"Human inspectors are very experienced and skillful, but can get tired and therefore miss certain defects. An automated system can avoid this pitfall, but eyes and brains process information very quickly.
"To be viable, an inspection system needs to balance issues of accuracy and speed."
Professor Nagarajah added inspection systems also need to be flexible enough to cope with different vehicle models and part variations.
"This can be further complicated because slight variations in part shape or location do not necessarily signify a defect," he said.
For example, hoses may not always be the same shape. "They might be turned in or out, but still be OK. Any system has to be able to recognise that. It underlines the difficulty these systems have in assessing what is good or bad," said Professor Nagarajah.
By collecting and then analysing numerous images of the nine components under various environmental conditions, Swinburne University of Technology researchers have developed methods and associated techniques to accommodate these variations. The approach also uses artificial intelligence techniques to help the system make a decision.
Once outcomes from the ARC-funded project are implemented the research team will be able to more accurately gauge the capabilities of current systems and associated technologies.