vocamp / dronedata Goto Github PK

After looking through the design for W3C/OGC sosa and the geolink patterns, I think it might make sense to break the patterns up a bit. We would have a core part (sUAS-voc) like sosa that is modeled after a schema.org vocabulary and no logics and a (sUAS-ont or whatever we call it) that has the extended logics. I also am thinking that the core vocabulary would be general enough that any type of drone mission (non-scientific) would be able to use it. And then have a module that extends Payload into a scientific platform that would have the additional features needed for sensors and observations.

In order to build momentum/motivate for further effort we need a way of showing the value of this to data providers and end users (by which I just mean Scientists xyz not institutions /etc even)

So looking for ideas for minimal demonstration options;

1. Use SOSA and our basic model to represent one of Joe's platforms - 3DR solo plus camera - in RDF, alongside some associated imagery+metadata also in RDF? If the ESIP SPARQL end point can also render the images that could be a easy-ish small demo?

2. More effort: setup a autoingestion system s.t data can stream from a drone - be transformed - and have a dynamic query running that pushes out a graph/some graphic response. I'm thinking eg: real time averaging of temp values, or even better if we had 2 independent sources of data that were both queried and combined (drone plus online weather station/something)

An existing drone ontology that appears to be fairly complete is available at dronetology.net. It has alignments to the old SSNX. It is not really pattern oriented and doesn't contain anything as formal in terms of logics as the semantic trajectory pattern. It has some mereology as part of the ontology. I guess we need to explore/discuss how this effort relates to what we are proposing.