Description of the OSIRIS-REx Simulated Test Models bundle V1.0 =============================================================== Bundle Generation Date: 2021-12-22 Peer Review: 2022_Asteroid_Review Discipline node: Small Bodies Node Content description for the OSIRIS-REx Simulated Test Models bundle =================================================================== Overview of archive package This archive package contains the global digital terrain models (DTMs) used in testing the shape generating software for the OSIRIS-REx mission and results. It includes both the truth (source) models and the test (generated) models. Before launch of the OSIRIS-REx mission, the mission team had to demonstrate that the suite of software tools had the capacity to conduct science and navigation operations around the asteroid Bennu. As such, this mission had the requirement to test the shape modeling software, stereophotoclinometry (Gaskell, et al., 2008), to determine if it could generate a high enough quality DTM for the navigation team to use its products for operations. Why this data set exists In order to test the software, the Space Operations Center (SPOC) was tasked with conducting a suite of tests for the validation and verification of the SPC software. The team had to demonstrate through a suite of tests that the SPC software met the following criteria: The software did not have substantial software errors (bugs) The software could generate a DTM The navigation team could use the DTM for operations The imaging plan collected enough images for the DTM generation The procedures for running the DTM generation were documented A suite of software tests were conducted to ensure that the entire mission team would be able to both produce and use the DTM generated by SPC. In order to run these tests, test data was needed. This test data must be generated by the mission. Required pieces to conduct DTM generation test: - Truth model - Nominal flight profile - Simulated actual flight profile - Simulated reconstructed flight profile - Images taken according to mission plan The truth DTM (which is a synthetic model) is the cornerstone of any DTM generation test. First, it is simulating the object that is being tested. All test data will be based upon this core model. The truth DTM will be used to render the entire suite of images, which are the primary input for shape modeling. As such, it is the first and most important piece of input data. Second, because it is the truth model, it is the standard to which the modeled DTM is tested. For most DTM evaluations, two models are subtracted from each other and the deviation between heights are calculated. The nominal flight profile is the trajectory and pointing of the spacecraft (and thus the camera) that the mission is designed to fly. When the mission team generates the commands for burns of the spacecraft, it is to follow this flight profile. The actual flight profile is the trajectory and pointing that the spacecraft actually followed during the mission. Because there are always errors in the amount of deltaV imparted during a burn maneuver, the spacecraft will never exactly fly the nominal flight profile -- there will be deviations. Further, the actual flight profile cannot be fully determined because there is measurement error in the tools to determine the spacecrafts position, such as Doppler ranging, star tracker data, camera distortions and timing errors. A reconstructed flight profile is the profile that the navigation team, using all available data, believes that the spacecraft actually took. Not only is the reconstructed flight profile different than the nominal flight profile (due to errors in maneuver burns), it is different than the actual flight profile (due to positional measurement errors). Both the simulated actual flight profile and the simulated reconstructed flight profile must be artificially generated by the mission team to make realistic simulated testing data. The simulated actual flight profile would typically contain error based upon the expected error in "spacecraft control", i.e. when the spacecraft is told to impart a given amount of delta v, how accurate is that burn. The simulated reconstructed flight profile would typically be based upon the simulated actual flight profile with additional errors based upon "spacecraft knowledge," i.e. how accurately can we calculate where the spacecraft is. Typically control error is larger than knowledge error. For this misson data set, we did not generate a simulated actual flight profile but used the nominal flight profile as what the spacecraft flew. While it was a reduction in realism of the test data, it was expected that it would not significantly impact the quality of the generated shape model. The simulated reconstructed flight profile had the expected level of knowledge errors added into the profile. What is provided in this release Truth Models We have included the global truth models that were the bases for the rest of the simulated data. Each truth shape model contains a different level of detail, quality and input parameters. Generally, as the testing progressed, the level of sophistication and quality improved . These global truth models are provided in different quality levels: 1,600,000, 400,000, 100,000 25,000 vectors. These translate into ground sample distances (GSD) of 75cm, 150cm, 300cm and 600cm They will be provided in the default SPC format of implicitly connected quadrangles (ICQ) Testing Models We have included the resulting models from the F4 test (that covered Approach and Preliminary Survey) and F6 (that covered Approach, Preliminary Survey and Detailed Survey). It is in ICQ format and a GSD of 75cm (about 1.6M vectors). What will be provided in future releases In future releases, we will be releasing higher resolution truth DTMs. Even the largest global DTMs (1,600,000 vertices) are only able to describe the surface with a GSD of about 75cm, i.e. there is a vertex every 75cm on the surface. To complete the full mission testing, regional DTMs were needed at a GSD of 5cm over the entire surface. It was even necessary to produce 0.25cm GSD DTMs in specific areas of the object. Also in future releases, we will be releasing the synthetic imagery that was generated from the high resolution DTMs. We will also be releasing what the mission team had as the nominal flight profile before launch and the simulated reconstructed flight profiles used during testing. Caveats to the data user ======================== Shape 3 model has a small region that is missing data. It did not impact the test.