Multiple robot data formats will be supported.  We're working on it - no really!
  • Most of the STOQS system has been installed and configured at NPS.  Initial tests are satisfactory.
  • Once the workflow is established, our maturity goal is that developing export/conversion code for a new robot will likely take 1-2 work days to develop.
  • Once a given robot's outputs are configured, ongoing usage can be repeated as long as the robot's data-export configuration is stable.

Information Flow

Robodata information flow converts raw data into structured datasets, then provide queriable information for analysis.

  • Information flow for robodata missionsData types, metadata annotations and database structure are carefully defined.
  • Robots collect well-defined track and sensor telemetry, imagery, video etc.
  • Mission operators record narratives, goals, outcomes, image captions, etc.
  • Conversion software reads telemetry according to metadata definitions, building a well-structured NetCDF mission file.
  • STOQS server loads NetCDF into the campaign database of robot missions
  • End users can query the STOQS server for filtered results, plots and visualizations.
  • Operator narratives, imagery, videos and 3D models will be stored separately

Exemplar Student Database Project

Robodata Schema ModelOur challenge is getting multiple datasets and databases to work together in a complementary way, rather than building a "super database" or telling everyone how to do things in a single fashion.

Three NPS students performed a database course project to design and create a prototype database model demonstrating the potential collection, presentation and dissemination of meaningful data about robodata experiments.  This is our template for continued design.