R&T often print models to enable engineers and stakeholders to get a better idea of what they are working with. These models will often be used as "lego", where people can build or deconstruct installations to get a sense for how the design will work in real life.
We might be able to enhance the visual aspect of the models by using augmented reality (AR). To make this feature available to as many people as possible we can use their mobile phone as device, use a web app so no installations is required, and then use QR codes to tag the model to provide tracking for AR.
MVP
Use a mobile phone or a laptop as device
Use a web app to drive AR experience
Populate real world model with QR codes (paper stamps) and use them as points of interest
Hard coded data is Good Enough™, do not worry about storage systems for now
Single player mode is Good Enough™
Any real world object is Good Enough™, like a paper box, lunch box
Be able to identify type of real world object the user are looking at, and show info related to that type.
Example: Type "lunchbox" has 4 Points of Interest, each point has information that identify the object type as well as specific point information
Document the pros and cons of this approach, as well as any other related experience made, as a simple markdown file in this repo
Stretch goals if MVP works
Can we enhance the physical model with CGI models?
Add extra "physical" features to the model, like pipes, generators etc
Be able to toggle visibility of CGI models using touchscreen
Be able to walk around the real world object and see that the CGI models stay in place
Can we enhance the physical model with audio?
(Note: html5 audio api can be a real ¤#"!¤"% to work with at times, tears of bravery is ok)
Let each point of interest have it's own sound snippet
Provide a simple interaction for how and when the sound should play
Example: Use gaze controls to "face-click" the POI
Can the user set interaction between points of interactions?
Example: the user can stretch a powerline between POI-1 and POI-2
Azure AD app proxy
Authentication and authorization as we can easily filter on ad groups.
Can be implemented as a proxy container in front of the app container to simplify integration
erlendb
commented
4 years ago
R&T often print models to enable engineers and stakeholders to get a better idea of what they are working with. These models will often be used as "lego", where people can build or deconstruct installations to get a sense for how the design will work in real life.
We might be able to enhance the visual aspect of the models by using augmented reality (AR). To make this feature available to as many people as possible we can use their mobile phone as device, use a web app so no installations is required, and then use QR codes to tag the model to provide tracking for AR.
MVP
Example: Type "lunchbox" has 4 Points of Interest, each point has information that identify the object type as well as specific point information
Stretch goals if MVP works
Can we enhance the physical model with CGI models?
Can we enhance the physical model with audio? (Note: html5 audio api can be a real
¤#"!¤"%to work with at times, tears of bravery is ok)Can the user set interaction between points of interactions? Example: the user can stretch a powerline between POI-1 and POI-2
Suggested technologies
A-Frame JS
Web VR library which now also support WebXR
Docker
For hosting web app anywhere
Azure AD app proxy
Authentication and authorization as we can easily filter on ad groups.
Can be implemented as a proxy container in front of the app container to simplify integration
Omnia Radix
CICD and hosting
Mockup