Deep Reinforcement Learning for Autonomous Drone Navigation: A Review
Keywords:
Deep Reinforcement Learning, UAV Navigation, Autonomous Systems, Sensor Fusion, Obstacle AvoidanceAbstract
This study examines the growing importance of Deep Reinforcement Learning (DRL) as a transformative approach for autonomous UAV navigation in complex, dynamic, and uncertain environments. Traditional rule-based and model-driven control strategies struggle with real-time adaptation, high-dimensional sensory data, and unpredictable obstacles, highlighting the need for learning-based methods. DRL overcomes these limitations by integrating deep neural networks with reinforcement learning to enable drones to learn navigation policies directly from raw sensory inputs such as images, lidar, and depth data. The reviewed literature demonstrates significant advancements, including hybrid learning frameworks, improved state representations, dynamic reward designs, sensor-fusion architectures, and distributed DRL models that enhance stability, decision accuracy, and collision avoidance. Despite challenges such as high training cost, sim-to-real transfer gaps, and sensitivity to reward design, DRL continues to evolve as a critical technology for reliable, adaptive, and scalable UAV autonomy across diverse real-world scenarios.
References
Karunanithy K, Velusamy B, Prasanth A. Remote Assessments and Aerial Imaging Using UAV for Disaster Management and Precision Agriculture with Immediate Response–A Case Study. Drones for Transportation Logistics and Disaster Management. 2025 Nov 20:371-98.
Tian C, Cho Y, Song Y, Park S, Kim I, Cho SY. Integration of AI with artificial sensory systems for multidimensional intelligent augmentation. International Journal of Extreme Manufacturing. 2025 Mar 27;7(4):042002.
Li H, Zhang Q, Zhao D. Deep reinforcement learning-based automatic exploration for navigation in unknown environment. IEEE transactions on neural networks and learning systems. 2019 Aug 6;31(6):2064-76.
Buyuktanir T, Aktas MS. On the Use of Embedding Techniques for Modeling User Navigational Behavior in Intelligent Prefetching Strategies. Concurrency and Computation: Practice and Experience. 2025 Feb 1;37(3):e8356.
Fujimura K. Motion planning in dynamic environments. Springer Science & Business Media; 2012 Dec 6.
Wu J, Wu QJ, Chen S, Pourpanah F, Huang D. A-TD3: An adaptive asynchronous twin delayed deep deterministic for continuous action spaces. IEEE Access. 2022 Dec 2;10:128077-89.
Dieter TR, Weinmann A, Jäger S, Brucherseifer E. Quantifying the simulation–reality gap for deep learning-based drone detection. Electronics. 2023 May 11;12(10):2197.
Lee EA. Determinism. ACM Transactions on Embedded Computing Systems (TECS). 2021 May 29;20(5):1-34.
Mieloszyk J, Tarnowski A, Tomaszewski A, Goetzendorf-Grabowski T. Validation of flight dynamic stability optimization constraints with flight tests. Aerospace Science and Technology. 2020 Nov 1;106:106193.
Dana KJ, Van Ginneken B, Nayar SK, Koenderink JJ. Reflectance and texture of real-world surfaces. ACM Transactions On Graphics (TOG). 1999 Jan 1;18(1):1-34.
Thi HL. Advancements in Quantum Computing: Revolutionizing Sensor Fusion and Data. Intelligent Sustainable Systems: Selected Papers of worlds4 2024, Volume 4. 2025 Jul 1;1180:123.
Wu J, Ye Y, Du J. Autonomous drones in urban navigation: Autoencoder learning fusion for aerodynamics. Journal of Construction Engineering and Management. 2024 Jul 1;150(7):04024067.
Houliotis K, Oikonomidis P, Charchalakis P, Stipidis E. Mission-critical systems design framework. Advances in Science, Technology and Engineering Systems Journal. 2018 Mar 12;3(2):128-37.
Luan T. A comprehensive review of simulation technology: development, methods, applications, challenges and future trends. International Journal of Emerging Technologies and Advanced Applications. 2024 Jun 24;1(5):9-14.
Sheng Y, Liu H, Li J, Han Q. UAV Autonomous Navigation Based on Deep Reinforcement Learning in Highly Dynamic and High-Density Environments. Drones. 2024;1(2):12–20.
Samma, Hussein SE-F. Autonomous UAV Visual Navigation Using an Improved Deep Reinforcement Learning. IEEE Access. 2024;2(June):79967–77.
Kalidas AP, Joshua CJ, Quadir A, Basheer S, Mohan S, Sakri S. Deep Reinforcement Learning for Vision-Based Navigation of uavs in Avoiding Stationary and Mobile Obstacles. Drones. 2023;1(2):5–14.
Doukhi O, Lee DJIN. Deep Reinforcement Learning for Autonomous Map-Less Navigation of a Flying Robot. IEEE Access. 2022;10(August):82964–76.
Hodge VJ, Hawkins R, Alexander R, Hodge VJ, Hawkins R. Deep reinforcement learning for drone navigation using sensor data. Neural Comput Appl [Internet]. 2021;33(6):2015–33. Available from: https://doi.org/10.1007/s00521-020-05097-x
Guo T, Jiang N, Li B, Zhu X, Wang Y, Du W. UAV navigation in high dynamic environments : A deep reinforcement learning approach. Chinese J Aeronaut [Internet]. 2021;34(2):479–89. Available from: https://doi.org/10.1016/j.cja.2020.05.011
Azar AT, Koubaa A, Mohamed NA, Ibrahim HA, Ibrahim ZF, Kazim M, et al. Drone Deep Reinforcement Learning : A Review. Electron Rev. 2021;1(2):1–30.
Anwar A, Raychowdhury A. Autonomous Navigation via Deep Reinforcement Learning for Resource Constraint Edge Nodes Using Transfer Learning. IEEE Access. 2020;1(2):1–9.
Muñoz G, Barrado C, Çetin E, Salami E. Deep Reinforcement Learning for Drone Delivery. Drones Artic. 2019;1(3):3–9.
Chao Wang, Jian Wang, Xudong Zhang And XZ, Department. AUTONOMOUS NAVIGATION OF UAV IN LARGE-SCALE UNKNOWN COMPLEX ENVIRONMENT WITH DEEP REINFORCEMENT LEARNING Chao Wang , Jian Wang , Xudong Zhang , And Xiao Zhang. Glob 2017. 2017;1(2):858–62.
Downloads
How to Cite
Issue
Section
License
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
