Virtual Reality for Simulating Autonomous Deep-Space Navigation and Mining

Abstract

In accordance with the space exploration goals declared by the National Aeronautics and Space Administration (NASA) in 2010 and 2013, the investigation of the deeper solar system becomes a central objective for upcoming space missions. Within this scheme, technologies and capabilities are developed that enable manned missions beyond low-Earth orbit - to lunar orbit, lunar surface, or even Mars and beyond. Particularly interesting targets are asteroids. They can serve as test beds for hardware and technology demonstration, which is needed prior to those aspired long-term missions. Asteroids can frequently be reached with smaller energy demands than those required for a mission to Moon or Mars. Furthermore, they are assumed to contain significant amounts of water and valuable metallic volatiles, which could serve as in-situ supplies for life support systems or spacecraft maintenance. Despite these technical facts, asteroids are also very interesting targets from a scientific point of view: They are remainders of the early formation phase of the solar system and are hold responsible for bringing life to Earth [DFJ90]. As the trend in future space exploration tends to focus on objects in deep space, the importance of autonomy increases on-board of spacecraft.With augmenting signal travel time due to great distances to Earth, it is difficult or even impossible to be able to react from ground on unexpected events for which time is a crucial factor. Up to this date, spacecraft in orbit follow specific timeline procedures during time-critical mission phases or pre-designed protocols in case unknown failures occur. The most common reaction on faults is the safe mode, during which the spacecraft shuts down every on-board module except the vital systems and awaits further (recovery) instructions from Earth ground stations. Hence, the demand for closed loop decision-making processes that are independent of the tele-commanding from ground. This includes not only the handling of errors but also navigation, guidance, and attitude/orbit control tasks. Therefore, the focus of this project is to make the spacecraft independent from the ground station as much as possible. This shall be achieved by autonomous navigation and autonomous decision making, so that it can determine optimal trajectories during flight and potential target asteroids autonomously for mining. The autonomy of the spacecraft is based on cognitive and biology-inspired algorithms. Assessment of these algorithms is necessary before they are applied in real scenarios. Therefore, algorithms have to be tested in a virtual environment with different virtual scenarios. This virtual environment should simulate motion of planets and asteroids, gravity, solar pressure, sensors of spacecraft, features of the asteroid, collision detection between asteroid and spacecraft for landing, etc. in real-time. In order to interact with this virtual environment, different 3D interaction metaphors have to be defined so that the user can change physical parameters, visualize different data, create different mission scenarios, change the spacecraft parameters, and even create new asteroid clusters and shapes (generated via 3D procedural modelling), which is necessary as the spacecraft might encounter new unknown asteroids.