There is a growing demand for unmanned aerial vehicles (UAVs) in the industry as they are being widely used in various areas such as healthcare, security, military missions, agriculture, etc. However, the increase in the production and use of UAVs requires the improvement of solid decision-making principles, safety, security, and relevant technologies. In this regard, the present study investigated the performance of different machine learning models in detecting faults and attacks in UAV systems. To achieve this, we systematically compared eight supervised models applied to the early detection of attacks and faults in the physical components of UAVs. To reach this purpose, the relative performances of each model are evaluated in two controlled testing scenarios.
Supervised models for detecting GPS attacks and faults in UAVs: a comparative analysis
2024-11-11
306336 byte
Conference paper
Electronic Resource
English