Aircraft inspection - is there a role for robots?

Citation

Willsher, S.J. (1998), "Aircraft inspection - is there a role for robots?", Industrial Robot, Vol. 25 No. 6. https://doi.org/10.1108/ir.1998.04925faa.002

Aircraft inspection - is there a role for robots?

Aircraft inspection ­ is there a role for robots?

The authorStephen J. Willsher is with the Structural Materials Centre, Defence Evaluation and Research Agency, Farnborough, UK.

Keywords Aircraft, Inspection, Robots

Robotic manipulators may provide the key to a leap forward in aircraft serviceability ­ by automating labour-intensive inspection. Traditionally non-destructive testing (NDT) techniques have been divided into two categories: spot measurements and single-shot images. In general ultrasonic, eddy-current and mechanical impedance methods were in the spot-measurement category and radiography in the single-shot imaging category. However, it is the ability to scan a spot-measurement device over an area and build up an image of the structure that has introduced a third category, commonly referred to as C-scanning. This is where robots may be able to automate and speed up some complex manual inspections.

The requirement to inspect large areas of structure as quickly as possible has been one of the NDT world's "holy grails" for many years. Much of the research into improved NDT techniques has focused on the development of new improved single-shot techniques such as laser shearography and thermography. However, a revolution has occured following the introduction of in-situ scanning systems that enhance the capabilities of the spot measurement techniques and potentially this is where robots could provide the ultimate scanning system.

So, how important a role does NDT play in the aerospace industry? During their lifetime, aircraft structures are subject to fatigue phenomena which have led to design lifing principles such as the "safe-life" and "fail-safe" concepts. More recently the "damage-tolerant" concept has been adopted in which it is accepted that undetectable defects from the manufacturing process can exist within the structure. An inspection plan is then implemented to detect any growing defects before they reach a critical stage.

One problem with conventional design lifing principles is the difficulty of predicting, or taking account of, corrosion in the airframe. This is the "ageing-aircraft" problem and was highlighted in 1988 when a Boeing 737 operated by Aloha Airlines suffered a major failure of the fuselage structure due to multi-site fatigue cracking in a corroded lap-joint. This resulted in much effort being put into developing techniques to detect corrosion at an early stage, particularly in multi-layered fuselage structure.

In addition, the increased use of composite materials in modern aircraft structures has introduced new problems for the non-destructive testing engineer. There is great concern over the phenomenon known as "barely visible impact damage" (BVID) in carbon-fibre composite. Composite materials exhibit excellent tensile strength and modulus, however they are vulnerable to low-energy impact. Just the act of dropping a tool on a composite surface can lead to delamination damage that may be undetectable by eye and which could then grow to a critical size under subsequent loading.

Before the widespread introduction of robots into the aircraft inspection world there are some obstacles that need to be overcome. Large, fixed automated and robotic scanning systems have been used by the aircraft manufacturers for some time. These include systems for the repeat inspection of large numbers of components such as turbine blades, discs and aircraft wheels, as well as automated ultrasonic immersion and water-jet scanners for large-area C-scan mapping. These systems, while ideal for post-manufacture inspection, tend not to be of interest to the aircraft operators owing to the restrictions in the number of components and aircraft types that the robot can be applied to, as well as the initial capital outlay. So the real challenge for robot developers is in the end-user market, i.e. the aircraft operators, both civil and military, and aircraft maintenance organisations. Fixed systems are not feasible as the operator, usually flying from more than one airport or air station, needs to take the inspection system to the aircraft. The initial financial outlay means that the operator will want to apply the system to more than one aircraft type. Crawler-type robots appear to be the most versatile in this respect.

At present no single, all-encompassing, high resolution NDT technique can detect all defect types on an aircraft structure. The robot should therefore be able to carry sensors from more than one type of instrument, whether it is an optical system, an eddy current test set, an ultrasonic flaw detector or a mechanical-impedance analyser. It should also be possible to map the inspection data as a function of the robot position. Aircraft operators are very interested in data fusion concepts with the intention that a map of defects throughout the airframe can be collated over the life of the aircraft. This is particularly important in the light of current trends for aircraft life-extension programmes in which the operator favours the introduction of extensive maintenance and re-building of their 20- or 30-year old airframes rather than buying new aircraft.

Despite the changes in emphasis on the requirements for NDT in the aerospace world it is clear that NDT technology is here to stay. While there are some interesting advances in other large-area scanning techniques, robotic systems with their speed and repeatability have an important place alongside this development.

The views expressed here are those of the author and are not necessarily those of the Defence Evaluation and Research Agency.