This academic survey provides a comprehensive overview of Federated Learning (FL), a distributed machine learning approach allowing collaborative model training without centralizing sensitive data. It details FL's architecture and communication protocols, highlighting key stages like local training and model aggregation. The text emphasizes technical challenges such as handling diverse data distributions, managing hardware differences, and reducing communication needs, along with privacy mechanisms like differential privacy. Furthermore, it discusses emerging research trends, real-world applications across various sectors like healthcare and finance, and outlines future research directions for this rapidly evolving field.