Conclusions
By observing the above results, several key takeaways can be derived. These observations
help to define the expected precision and accuracy of the TCH model viewer, while providing an
opportunity to prescribe best practices which optimize the user's experience. When assessing
scale error it was found that the XR10 projected holograms at a near perfect 1:1 ratio, exhibiting
an error of only -0.03%. However, due to the unique characteristics of each device, model, and
user, scale error determined in this study should only be taken as a general guidance.
In completing the remainder of this evaluation, two things were clear: (1) drift error causes the
greatest impact on model accuracy and reliability, and (2) WLT significantly diminished this
impact. Comparing the observed model coordinates to their expected locations highlighted a
decrease in error of 3x, shrinking from 1.14% to 0.37% between WLT and non-WLT solutions.
Similarly, when comparing the two solutions via baseline, it was found that drift was reduced by
4x, shrinking from 1.27% to 0.31%. Over a distance of 50m this is a staggering improvement of
nearly 0.4m (1'3") in error.
Overall, it is clear that the World Locking
Tools engine has significantly improved
the precision and accuracy of models
within the TCH platform. With this
knowledge, it is now easier than ever to
confidently put mixed reality to work on
full scale construction projects, helping
to ensure projects come in on time and
in budget.
Best Practice Recommendations
To mitigate the effects of scale and drift errors, users should keep the radius of their
mixed-reality work area limited to less than 50m (160'), with an optimal working radius of <30m
(100'). If the work environment cannot be confined to such a radius, users should ensure they
re-anchor their models intermittently as they travel beyond this optimal zone. This
recommendation may shift depending on a user's individual precision and accuracy needs,
dependent on their individual workflows.
17
