Intelligent Path Navigation for Autonomous Drones Using Deep Neural Networks
Publication Date : 24/03/2025
Author(s) :
Volume/Issue :
Abstract :
Path planning plays a critical role in autonomous navigation, particularly in environments filled with multiple obstacles. Effective path planning ensures that an autonomous system can reach its destination efficiently while avoiding collisions. This study focuses on the Sequential_9 model, a navigation approach designed to optimize both efficiency and safety. The model was tested in a scenario containing five obstacles, where it successfully computed an optimized route while maintaining safe distances from potential hazards. The evaluation results demonstrate the model’s remarkable performance. The total path length taken by the Sequential_9 model was 75.18 units, which is extremely close to the straight-line distance of 75.12 units, leading to a path efficiency of 99.92%. Such high efficiency indicates that the model minimizes unnecessary deviations while navigating through complex environments. Moreover, the model ensured a minimum clearance of 9.16 units from obstacles, highlighting its strong collision avoidance capabilities. These findings underscore the Sequential_9 model’s ability to generate precise, safe, and near-optimal paths, making it a promising solution for applications requiring autonomous movement in dynamic and constrained spaces. Future work may involve testing under varying obstacle densities to further validate its robustness and adaptability. Keywords: path planning, autonomous navigation, collision avoidance, Sequential_9 model, path efficiency, safety margin.
No. of Downloads :
0