Over the past few decades, automated aerial vehicles have been essential in the upgradation of fifth-generation wireless communication. These aerial vehicles provide us with many enhancements in various fields, including digital agriculture, salvage detection, variable detection, cost-efficiency, versatility, and spectral efficiency. At the same time, they are also facing some harsh challenges that is related to security and low power management Additionally, UAVs equipped with sensors and capturing devices, which capture large amounts of data and information, can contain specific information where security is a major concern. Thus, the subset of artificial intelligence, i.e., the use of machine learning algorithms like regression, reinforcement learning, and deep reinforcement learning techniques can be used to optimize and conserve the discharged power and improved security reasons. Anomaly detection techniques are vital in identifying and managing security threats. These methods are very important for probing email communications to detect anomalous or suspicious activities, which helps maintain endpoint security through robust encryption measures that block potential threats directly. Additionally, anomaly detection is key in analyzing network traffic to identify and prevent distributed denial-of-service (DDoS) attacks, thereby bolstering the overall security framework. In this work, we review the key research works done in the area of anomaly detection for UAVs.