Mars Rover Image Data Prioritization for Increased Mission Science Return

Rover traverse distances are increasing at a faster rate than downlink capacity is increasing. As this trend continues, the quantity of data that can be returned to Earth per meter of traverse is reduced. The capacity of the rover to collect data, however, remains high. This circumstance leads to an opportunity to increase mission science return by carefully selecting the data with the highest science interest for downlink. We have developed an onboard science analysis technology for increasing science return from missions. Our technology evaluates the geologic data gathered by the rover, and prioritizes this data for transmission, so that the data with the highest science value is transmitted to Earth. Although our techniques are applicable to a wide range of data modalities, our initial emphasis has focused on image analysis, since images consume a large percentage of downlink bandwidth. We have further focused our foundational work on rocks. Rocks are among the primary features populating the local Martian landscape. Characterization and understanding of rocks on the surface is a first step leading towards more complex in situ regional geological assessments by the rover. Data prioritization involves two processes: the identification of significant features in the data and the use of these features to assess the scientific value of the data. In our current application, we locate rocks in the image data and then extract properties of each rock, including albedo, visual texture and shape. These properties are then used to prioritize the rocks and thereby prioritize the images of the rocks. Three prioritization methods have been developed: identification of key target signatures, novelty detection, and sampling representative rocks. The use of these three methods ensures that three exploratory science objectives are met. First, objects known to be of very high interest, such as indicators of water, will be immediately recognized if encountered. Second, unexpected objects that may lead to key discoveries will be noted. It is, however, also important to have an understanding of the typical characteristics of the region. Our final prioritization method selects the most representative rocks for the downlink queue. As NASA continues to increase the number of high data volume missions simultaneously operating, an onboard mechanism for the prioritization of data tagged for downlink that can increase the science content returned for a fixed bandwidth will be invaluable to scientists who will continue to compete for downlink time.