In contemporary times, the utilization of drone detection systems has found extensive application in real-life scenarios. The necessity for real-time drone identification, particularly in defence, has prompted the investigation of new technologies for increased surveillance. This research paper delves into the realm of drone detection using a repertoire of machine learning models, including Naïve Bayes (NB), Random Forest (RF), Logistic Regression (LR), Support Vector Machine (SVM), Convolutional Neural Network (CNN), and Recurrent Neural Network (RNN). This paper evaluates the efficacy of these models in detecting drones in different datasets, emphasizing their significance in practical scenarios. This research employs a comprehensive set of machine learning models to tackle the intricate challenge of drone detection. After rigorous testing on diverse datasets, the findings reveal notable accuracy rates for the employed machine learning models. Specifically, the RNN model emerges as the most effective, achieving accuracy rates of 93%, 99%, 99%, 99%, and an exceptional 99.85% across distinct datasets. This underscores the prowess of RNN in addressing the complexities of real-time drone detection.