Mission Design for Combined Lander-rover Modeling of a Skylight

Introduction: Combined lander-rover modeling is the transformational means for developing high resolution, color, three-dimensional representations of planetary features such as cave entrances and skylights. Lander-rover modeling combines registered overflight imagery with rover-based surface imaging techniques to build highly accurate co-registered models. The models produced by this approach exhibit high resolution from surface modeling, and high accuracy resulting from coregistration of surface and overflight models. Such models require specialized trajectories designed to provide high precision relative to the selected feature while supporting the requirements for safe landing. The architecture detailed here combines lander flyover with extended investigation by robotic rover. Real-time data from cameras and LIDAR are combined with existing satellite imagery to navigate precisely to a selected landing zone, identify a safe landing location, and maneuver past hazards to safely touch down. While flyover provides birds-eye views of the feature, landing views are limited by fuel constraints. The rover provides low-angle, detailed views of specific areas of high interest detected from above. Rover and lander data are combined in post-processing to determine landing location to within 5m accuracy and to reconstruct the actual landing trajectory and feature models with 10cm precision. A specific case study of a Lunar skylight known as the Marius Hills Hole[1, 2] is detailed. Targeting Skylights: Skylights are excellent candidates for the next generation of planetary missions. These accesses to lava tubes and expansive caves exist on planetary bodies throughout the solar system. Caves and lava tubes are safe havens that will protect astronauts from extreme temperatures, micrometeorite impacts, and radiation. Subsurface caverns also preserve unique geologic environments. However, skylight missions present challenges well beyond those of conventional missions. Landing zones near skylights are typically hazardous precluding standard blind landing techniques. The tunnels themselves are too dangerously unknown to initiate exploration with humans. Rather, robotic landers will precisely land near these sites and deploy roving explorers to map and characterize these destinations for future human missions. Communication delay during landing, and complete communication blackout during underground operation demand unprecedented levels of autonomy above and below the surface. Development of highly accurate models requires new techniques for data fusion and localization. Figure 1: Trajectory for combining flyover and surface modeling of a skylight. Top: landing trajectory (green) with LIDAR views (red) of the skylight. Bottom: rover path (blue) overlaid on lander trajectory (green) as the rover circles around the skylight.