Flyover Modeling of Planetary Pits – Undergraduate Student Instrument Project

Abstract:

On the surface of the moon and Mars there are hundreds of skylights, which are collapsed holes that are believed to lead to underground caves. This research uses Vision, Inertial, and LIDAR sensors to build a high resolution model of a skylight as a landing vehicle flies overhead. We design and fabricate a pit modeling instrument to accomplish this task, implement software, and demonstrate sensing and modeling capability on a suborbital reusable launch vehicle flying over a simulated pit. Future missions on other planets and moons will explore pits and caves, led by the technology developed by this research.

Sensor software utilizes modern graph-based optimization techniques to build 3D models using camera, LIDAR, and inertial data. The modeling performance was validated with a test flyover of a planetary skylight analog structure on the Masten Xombie sRLV. The trajectory profile closely follows that of autonomous planetary powered descent, including translational and rotational dynamics as well as shock and vibration. A hexagonal structure made of shipping containers provides a terrain feature that serves as an appropriate analog for the rim and upper walls of a cylindrical planetary skylight.

The skylight analog floor, walls, and rim are modeled in elevation with a 96% coverage rate at 0.25m² resolution. The inner skylight walls have 5.9cm² color image resolution and the rims are 6.7cm² with measurement precision superior to 1m.

The multidisciplinary student team included students of all experience levels, with backgrounds in robotics, physics, computer science, systems, mechanical and electrical engineering. The team was commited to authentic scientific experimentation, and defined specific instrument requirements and measurable experiment objectives to verify successful completion.This work was made possible by the NASA Undergraduate Student Instrument Project Educational Flight Opportunity 2013 program. Additional support was provided by the sponsorship of an IMU by KVH industries and mentorship was provided by members of the NASA ALHAT team. In addition to external support, this project was supported by two CMU Small Undergraduate Research Grants, the availability of a high-power CMU LIDAR sensor, dedicated workspace, and mentorship from research and shop faculty.