SPATIAL DATA INFRASTRUCTURE FOR MARS ROVERS

Planetary missions require a spatial data infrastructure (SDI) that unites spacecraft engineering data with science instrument results in an accessible interface. We’ve developed a multi-mission geographic information system (MMGIS) composed of programs to find the 3D location of science data products on a planetary surface. Our first test case is for Mars rovers, though our goal is to provide the same SDI for lander, sub-orbital (e.g. ‘drones’), and orbital platforms. Our goal is to provide a SDI to access science data in their geospatial position so mission team members can quickly assess new results and make decisions for future observations.

Mars rovers have three types of instruments: ‘fixed body’ that are pointed according to the orientation of the rover (e.g. MARDI descent camera), ‘mast’ that use azimuth and elevation to target objects (e.g. MASTCAM or CHEMCAM), and ‘arm’ which are instruments deployed on the end of an extendable arm and capable of reaching the surface (e.g. MAHLI microscopic imager). Using the rover position and orientation, localization programs can interpolate the northing, easting, and elevation for each instrument type using the instruments known location on the rover and its pointing parameters. For ‘arm’instruments, the inverse-kinematics are calculated to locate the instrument, while for ‘mast’ instruments, the pointing is used to find the surface intersection of the observation via ray casting.

The science product geospatial information is compiled into a database and presented in a webGIS using open source javascript mapping and graphing libraries to present a tripartite view linking an insitu view, traditional 2D map, 3D presentations of science data. The views allow display of targeted features on image panoramas as well as viewing sub-surface features. Taking advantage of a web-based platform, no software needs installation by the mission user community and is updated automatically on the server side for all users. The SDI is ‘serverless’ in that it doesn’t rely on a ‘geoserver’ to display images or vector data. Raster data is compiled into Tile Map Service (TMS) tiles and GeoJSON for vector data. Raw data files can be sampled to create elevation cross-sections or raster classifications like slope. Testing and deployment is ongoing with current and future rover missions.