Obstacle avoidance in aerial robotics remains a critical challenge, particularly in environments with uncertain terrain and weather conditions. This study introduces a Constrained Q-Learning model that leverages spatio-temporal data from LiDAR and OctoMap to achieve Zero-Shot Execution (ZSE) for autonomous Unmanned Aerial Vehicle (UAV) navigation in unseen environments, eliminating the need for iterative training. Experimental evaluations are conducted using a high-fidelity simulator across three environments: random forests, clustered forests, and metropolitan areas, under varying obstacle densities and flight velocities. The proposed model demonstrates a 100% success rate, achieving average flight times of 45 s for slow velocities (below 2.5 m/s) and 34 s for fast velocities (above 2.5 m/s). Comparative analysis with 90 Human-to-Computer (HTC) flights (slow and fast velocities), conducted by three pilots under identical conditions, shows the proposed model reduces flight time by 33% (1.5 times faster) while enhancing path optimization. Additionally, the model matches the path selection efficiency of a standard Q-Learning approach without requiring iterative training, highlighting its robustness and scalability for autonomous UAV navigation in complex environments and GPS-denied locations.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Autonomous Navigation in Swarm of UAVs Using Spatio Temporal Data and Constrained-Reinforcement Learning


    Additional title:

    Communic.Comp.Inf.Science



    Conference:

    International Conference on Engineering Applications of Neural Networks ; 2025 ; Limassol, Cyprus June 26, 2025 - June 29, 2025



    Publication date :

    2025-06-22


    Size :

    14 pages





    Type of media :

    Article/Chapter (Book)


    Type of material :

    Electronic Resource


    Language :

    English






    Supporting UAVs Swarm Missions by Multi-Agent Reinforcement Learning

    Fusco, P. / Porcelli, L. / Palmieri, F. et al. | IEEE | 2025


    Spatio‐temporal dynamic navigation for electric vehicle charging using deep reinforcement learning

    Ali Can Erüst / Fatma Yıldız Taşcıkaraoğlu | DOAJ | 2024

    Free access

    Spatio‐temporal dynamic navigation for electric vehicle charging using deep reinforcement learning

    Erüst, Ali Can / Taşcıkaraoğlu, Fatma Yıldız | Wiley | 2024

    Free access