Computer Graphics Support in Head-Mounted Displays for Helicopter Guidance

Abstract

Providing navigational information from sensors and databases to a rotorcraft pilot on Head-/Helmet-Mounted Displays during Degraded Visual Environment situations is a challenging problem in computer graphics and a substantial task in avionics engineering. Data from different sensors such as GPS and LiDAR sensors as well as transponders needs to be processed in order to provide more high-level descriptions and symbols than raw data. Subsequently, the sensor data needs to be not only displayed but also abstracted dependent on the flight situation during a mission according to the pilot’s needs for spatio-temporal coherence. This dissertation investigates the use of a local-to-global method to create digital terrain models for LiDAR data classification as an extension to existing local approaches optimized for avionic hardware. Furthermore, a clustering approach as pre-processing as well as different methods for rendering of non-classified Li- DAR data are analyzed and provided within a flight simulator environment. We present an efficient 3D meshing approach that avoids cluttering of rendered elements in the helmet-mounted display. This meshing approach is combined with even more abstract renderings at different Levels-of-Detail. Similarly, we apply a density-based cluster method on traffic information in order to create a spatiotemporal coherent 2D abstraction with minimal impact to the pilot’s workload. These methods are completed with an alternative landing site selection as well as a terrain rendering approach motivated by automatically created artistic stylizations. The results of this dissertation are advantageous for navigation through different flight situations such as landing and flights in low altitude.We extend state-ofthe- art methods already applied in flight decks for operational use. The resulting visualizations are prototypes that have been analyzed in a small user study with a professional helicopter pilot. More exhaustive user studies with experienced pilots reflect the subsequent step. which is left for future work.