Robust and Reliable Decision-making for Unmanned Aerial Vehicles

Abstract

Machine learning-based models for object detection rely on large datasets of labeled images, such as COCO or ImageNet. When models trained on these datasets are applied to aerial images from Unmanned Aerial Vehicles (UAVs), the problem arises that the conditions under which the training images where created (for example, light, altitude, or angle) may be different in the environment where the UAVs are put into practice, leading to failed detections. This problem becomes even more pressing when it comes to safety critical applications where failures can have huge negative impacts and also constitutes an obstacle for certification of cognitive components in UAVs. This talk will show that using, both, artificial andreal images for training has a positive effect on the performance of object detection algorithms. Since simulated images are easy to create and object labels are inherently given, the presented approach shows a promising direction for scenarios where adequate datasets are difficult to obtain, as well as for targeted exploration of weak points of object detection algorithms.