Multi-Task Regression-based Learning for Autonomous Unmanned Aerial Vehicle Flight Control within Unstructured Outdoor Environments

Increased growth in the global Unmanned Aerial Vehicles (UAV) (drone)industry has expanded possibilities for fully autonomous UAV applications. Aparticular application which has in part motivated this research is the use ofUAV in wide area search and surveillance operations in unstructured outdoorenvironments. The critical issue with such environments is the lack ofstructured features that could aid in autonomous flight, such as road lines orpaths. In this paper, we propose an End-to-End Multi-Task Regression-basedLearning approach capable of defining flight commands for navigation andexploration under the forest canopy, regardless of the presence of trails oradditional sensors (i.e. GPS). Training and testing are performed using asoftware in the loop pipeline which allows for a detailed evaluation againststate-of-the-art pose estimation techniques. Our extensive experimentsdemonstrate that our approach excels in performing dense exploration within therequired search perimeter, is capable of covering wider search regions,generalises to previously unseen and unexplored environments and outperformscontemporary state-of-the-art techniques.