Research on passive location algorithm of UAV based on TDOA

In the airport environment, foreign body debris (FOD) can seriously damage the aircraft engine and harm human health, but the traditional method can not detect foreign body accurately and timely. Based on this, this paper proposes a method based on improved YOLOv8 to detect FOD. To solve the problem of slow detection of the original model, FasterBlock module is introduced in the backbone network to replace C2f module in the original network to speed up the model calculation. Aiming at the problem of low detection accuracy, DilateFormer module is introduced to realize multi-scale sensing. Furthermore, the loss function is modified to further improve the detection accuracy. The experiment on the airport runway foreign body detection data set shows that the improved algorithm mAP50 is 95.5%, which is 4 percentage points higher than the improved algorithm. The newly designed algorithm has a good application prospect in FOD detection with low model complexity and high detection accuracy.