Agrawal Sameer, Patle Bhumeshwar K, Sanap Sudarshan
Department of Mechanical Engineering, MIT Art, Design & Technology University, Pune 412201, India.
Sensors (Basel). 2025 Jan 31;25(3):863. doi: 10.3390/s25030863.
Autonomous aerial drone navigation is a rapidly growing topic of research due to its vast application in various indoor applications, including surveillance, search and rescue missions, and environmental monitoring. Current research focuses on the implementation of neighborhood dragonfly algorithms (NDAs) for path planning for single and multiple drones in various indoor environments containing stationary and moving obstacles. The collaborative behavior of dragonflies is a key concept in the current study that helps in exploring the solution space effectively and results in a faster convergence rate. To validate the performance of the proposed NDA approach, various environments are created in real time, and replicas of the same are generated using MATLAB software. Our analysis shows a close agreement between simulation and experimental results, with path length and navigational time differences of less than 5.7%. This underscores the consistency and feasibility of the NDA approach, placing the groundwork for robust and efficient drone navigation systems. The proposed NDA approach is also compared with those already developed, like IACO and PRM, in a similar environment. The NDA approach shows a better performance in terms of smooth path planning and path length optimization. The saving in path length is more than 5%.
自主空中无人机导航是一个快速发展的研究课题，因为它在各种室内应用中有着广泛的应用，包括监视、搜索和救援任务以及环境监测。当前的研究重点是在包含静止和移动障碍物的各种室内环境中，为单架和多架无人机的路径规划实施邻域蜻蜓算法（NDA）。蜻蜓的协作行为是当前研究中的一个关键概念，有助于有效地探索解空间，并实现更快的收敛速度。为了验证所提出的NDA方法的性能，实时创建了各种环境，并使用MATLAB软件生成了相同环境的副本。我们的分析表明，模拟结果与实验结果非常吻合，路径长度和导航时间差异小于5.7%。这突出了NDA方法的一致性和可行性，为强大而高效的无人机导航系统奠定了基础。在类似环境中，还将所提出的NDA方法与已开发的方法（如IACO和PRM）进行了比较。NDA方法在平滑路径规划和路径长度优化方面表现出更好的性能。路径长度节省超过5%。
