Efficient Rendering for Non-Gridded Data Sets and Procedural Terrain

An ability to render photorealistic images enables validation of Terrain Relative Navigation (TRN) algorithms in a fully simulated scenario where baseline truth is known exactly, and because the terrain is known exactly it can be used to automatically generate labelled training data for Machine Learning techniques. For descent and landing scenarios, images and LIDAR measurements must be simulated over a very large region (as the spacecraft descends from orbit) but must also be extremely high resolution as the spacecraft approaches the surface. Datasets of sufficiently high resolution often do not exist at all, and when they do are often too large to be loaded into memory all at once. This introduces difficulties with simulating images and LIDAR measurements efficiently. In this IRAD, we seek to continue development of an existing rendering tool to advance a dynamic Level-of-Detail system for managing extremely large scenes, as well as for generating procedural terrain, which will help address both of these concerns. These algorithms will be implemented into an ongoing development project at NASA Goddard Space Flight Center known as Vira, which is a high-fidelity ray tracing and rendering API.

These new algorithms would be packaged in the existing Vira rendering tool which is currently used for several ongoing projects. These implementations would be available to anyone developing Terrain Relative Navigation (TRN) algorithms, whether that be in industry or academia. It could allow for more accurate simulation of scenarios that are currently difficult to model accurately around many different planetary bodies. Additionally, the implementations developed here could be useful in other scientific contexts beyond just TRN, such as development of Hazard Avoidance (HA) algorithms, or even in testing mapping algorithms as it would be able to provide extremely high-resolution synthetic images where truth in the underlying topology is known exactly.

This is early/mid-stage (TRL 4) — the most realistic path in is NASA SBIR/STTR, which funds small businesses and research institutions to develop technology aligned with NASA's needs (equity-free, phased funding). Check whether a current SBIR/STTR solicitation topic overlaps with this project's technology area, or contact the project directly (above) to ask.

None of these are guaranteed paths for this specific project — TechPort itself doesn't have an "apply" button. Reaching out to the contact(s) above with a specific question is usually the fastest way to find out what's actually open.