An intuitive and unobtrusive method for visualizing and interacting with large multidimensional scan data using state of the art augmented reality technology.
From challenges to solutions
Modern approaches to quality control like non-destructive testing (NDT) produce an ever increasing amount of data such as high resolution datasets. Manual inspection can be very error prone and time consuming.
Use augmented reality devices to present NDT data sets in an intuitive and unobtrusive manner. Mapping the location-dependent measurement data with the geometry of the real component reduces the errors caused by manual mappings and increases inspection speed.
An ecosystem which makes it easily possible to prepare NDT data for visualisation on AR devices such as the Microsoft Hololens or smartphones (Android, ARCore). Including mechanisms for mapping this NDT data directly on the physical object.
The ever increasing need for non-destructive testing (NDT) – in particular in industries targeting „zero failure production“ – has lead to an abundance of high resolution multidimensional scan data sets. While these data sets enable a unprecedented level of quality in the component under test they do come with a challenge:
Visualizing and inspecting these data sets is a time consuming and largely non intuitive process – with the largest challenge being the registration of the scan data to the real world component.
In recent years the development of affordable and reliable virtual and augmented reality platforms has gained significant traction, primarily driven by the consumer electronics industry.
The idea of this research project is to utilize this relatively low cost platforms with the imaging capabilities of modern NDT measurement methods to provide an intuitive and accurate inspection workflow.
By mapping NDT scan results to the inspected component a co-registration can be achieved. This enables the overlay of location-dependent measurement data on the component and thus enables the operator of such a system to quickly inspect locations of the component both on the real component and on in the scan data set.
Furthermore by transforming the scan data in e.g. color coded thickness maps an overlay can be projected on the real world component and thus augment the perception of the operator and not only speed up the inspection process but potentially enable the detection of otherwise missed defect.
In order to prevent an vendor lock in in the rapidly changing and developing field of augmented reality platforms we developed an ecosystem that enables the quick adaption of both existing and upcoming augmented reality solutions – with current support ranging from Microsoft Hololens to Android / ARCore based platforms.
In addition mechanisms for mapping NDT data data directly on the physical object have been developed.