Simulating planetary approach and landing to test and verify autonomous navigation and guidance systems

PANGU v4 (Planet and Asteroid Natural scene Generation Utility) is a 64-bit software package designed to support testing and validation of vision-based and LiDAR-based landers and other spacecraft applications. PANGU generates simulated camera and LiDAR images of planetary surfaces and small bodies to test, verify and validate mission scenarios such as surface-relative navigation, approach and landing and surface rovers. PANGU models can be based on real or synthetic terrain elevation data and then artificially enhanced to increase the resolution of the model where required to meet the image pixel resolution requirements of the mission. Large models greater than 80 GB in size can be created to support complex scenarios subject to sufficient available memory resources. The custom Qt-based OpenGL renderer uses modern Graphics Processing Units (GPUs) to generate images for either open-loop or closed-loop testing at high-speeds to run simulations at near real-time (e.g. 10 HZ or faster).PANGU can generate realistic simulated images for a wide variety of planetary surface types and includes a parameterised camera model, atmospheric effects and Bidirectional Reflection Distribution Functions (BRDFs) such as Hapke, Oren-Nayar and Lunar Lambert, which can be tuned to match the reflection properties of different planetary surfaces. A user manual with tutorials, examples and case studies allows new and expert users to quickly create and run simulations.This paper describes the creation and rendering of PANGU planetary surface models and presents example scenarios to showcase the variety of potential missions that can be simulated.