当前位置: X-MOL 学术Transp. Res. Part A Policy Pract. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
IoD swarms collision avoidance via improved particle swarm optimization
Transportation Research Part A: Policy and Practice ( IF 6.3 ) Pub Date : 2020-11-24 , DOI: 10.1016/j.tra.2020.09.005
Gamil Ahmed , Tarek Sheltami , Ashraf Mahmoud , Ansar Yasar

Drones flights have been investigated widely. In the presence of high density and complex missions, collision avoidance among swarm of drones and with environment obstacles becomes a challenging task and indispensable. This paper aims to enhance the optimality and rapidity of three dimensional IoD path generation by improving the particle swarm optimization (PSO) algorithm. The improvements include using chaos map logic to initialize the population of PSO. Also, adaptive mutation is utilized to balance local and global search. Then, the inactive particles are replaced by new fresh particles to push the solution toward global optimal. Furthermore, Monte Carlo simulation is carried out and the results are compared with slandered PSO and with recent work CIPSO. The results exhibit significant improvement in convergence speed as well as optimal solution which prove the ability of proposed method to generate safety path for IoD formation without collision with terrain obstacle and among drones.



中文翻译:

通过改进的粒子群优化方法避免IoD群体碰撞

无人机的飞行已被广泛调查。在高密度和复杂任务的情况下,避免无人机群之间以及环境障碍下的碰撞成为一项艰巨的任务,而且是必不可少的。本文旨在通过改进粒子群算法(PSO)来提高三维IoD路径生成的最优性和速度。改进包括使用混沌映射逻辑来初始化PSO的填充。另外,利用自适应变异来平衡局部搜索和全局搜索。然后,将非活性颗粒替换为新的新鲜颗粒,以将解决方案推向全局最优。此外,进行了蒙特卡洛模拟,并将结果与​​诽谤的PSO和最新的CIPSO进行了比较。

更新日期:2020-11-25
down
wechat
bug