This paper provides a comprehensive review of intelligent path planning methodologies that have become important due to the increasing demand for autonomous systems in complex and dynamic environments. Graph-based navigation algorithms, linear programming techniques, and evolutionary computational methods have been used as fundamental approaches in this field for decades. Recently, deep reinforcement learning (DRL) has emerged as a powerful method to enable autonomous agents to learn optimal navigation strategies through their interactions with their environment. This paper provides a comprehensive overview of existing approaches and recent advances in DRL applied to path planning tasks, focusing on autonomous vehicles, drones, and robotic platforms. We categorize the main algorithms in both the existing and learning-based paradigms, and highlight their innovations and practical implementations. We then discuss their strengths and limitations in detail in terms of computational efficiency, scalability, adaptability, and robustness. Finally, we identify key open challenges and suggest promising directions for future research. We pay special attention to hybrid approaches that integrate DRL and classical planning techniques to take advantage of learning-based adaptability and deterministic reliability, suggesting promising directions for robust and resilient autonomous navigation.
