A Survey on AI-Based Detection Methods of GPS Spoofing Attacks on UAVs

Global Positioning System (GPS) spoofing attacks on unmanned aerial vehicles (UAVs) can cause severe consequences, including abnormal behavior, privacy breaches, safety hazards, financial losses, and national security risks. Hence, detecting and preventing GPS spoofing attacks on UAVs is paramount to ensure their safety and security. This survey paper explores the cutting-edge techniques based on artificial intelligence (AI) proposed in the literature for detecting GPS spoofing attacks on UAVs. We present an overview of the UAV navigation system and then delve into the concept of GPS spoofing, encompassing its various forms, such as soft and hard spoofing. Furthermore, we discuss the advantages of utilizing AI in developing detection methods compared to traditional approaches. Subsequently, we analyze the detection methods based on deep learning (DL) and machine learning (ML) proposed in the literature, evaluating their strengths and weaknesses whenever feasible. Our survey provides valuable insights for researchers and practitioners working in the fields of AI and UAV security with the target of developing more advanced and effective AI-based methods for GPS spoofing detection on UAVs.