Exploration in urban area

Especially near buildings, the navigation solution of a GNSS-based navigation system may be significantly degraded due to satellite shadowing and multipath propagation. Therefore, methods to improve the accuracy of the navigation solution by fusion of different sensors are developed. Furthermore, the increase of the availability and reliability of the navigation solution by means of navigation filters and integrity checks adapted to different situations is in the focus of research interest. This enables the autonomous flight of MAVs in difficult situations without the need for human intervention.

Autonomous exploration of indoor environments

Reconnaissance of buildings by unmanned aerial vehicles can provide rescue workers with important information outside their own field of vision, thus avoiding unnecessary, incalculable risks for first responders. In order to successfully carry out such reconnaissance missions, the MAV must be able to independently explore all accessible interior spaces in unknown environments, react flexibly to obstacles and then return safely to the ground station. This presents the micro-helicopters, which are limited in computational capacity and payload, with a variety of dynamic challenges to which classical systems are often unable to respond in real time. Therefore, a multi-sensorial system for autonomous exploration is being developed, including in particular reactive strategies for safe and efficient indoor operations - i.e. without GPS reception. Onboard processing of heterogeneous sensor data is intended not only to enable comprehensive exploration, but also to enable the aircraft to navigate in a robust and situationally aware manner. Mission-specific acquired data can then be directly transferred into e.g. building plans for rescue forces.

Mapping of interior spaces

With the help of various sensors, in particular the laser range finder, the geometry of the environment can be mapped with simultaneous self-localization. Such metric maps are suitable for navigation of the aircraft, but less so for interpretation of the underlying scene. Therefore, the metric map is augmented with mission-specific qualitative features. The resulting map enables on the one hand the aircraft to recognize the situation and on the other hand the rescue forces to better plan the mission.