北京交通大学自动化与智能学院

侯涛刚课题组论文发表于《IEEE International Conference on Unmanned Systems (ICUS)》

发布于 2023-11-20

    我所侯涛刚老师课题组完成的论文《Improved Multi-Strategy Salp Swarm Algorithm for UAV Path Planning》于2023-10在《IEEE International Conference on Unmanned Systems (ICUS)》正式发表。论文摘要如下:

Abstract: In response to the unmanned aerial vehicle (UAV) path planning in a three-dimensional environment map, an improved multi-strategy salp swarm algorithm (IMSSA) is proposed under the condition of the environmental state. Based on the original salp swarm algorithm, introduce a Piecewise chaotic map to initialize the population, and use the Levy flight strategy to update the leader's position preventing the algorithm from falling into local optimum and enhancing the search capabilities; add a non-uniform Gaussian mutation operator in the follower's position update stage, that not only improves the search accuracy but also avoids individuals falling into local optimum. This paper's improved multi-strategy salp swarm algorithm is applied to solve the UAV path planning problem in a three-dimensional space environment. First of all, a three-dimensional environment model is established according to the grid method. Furthermore, considering UAVs' mobility efficiency, safety avoidance, and movement coherence, the mathematical model of UAV motion is determined, and the path planning problem is transformed into optimizing the cost functions. At last, the IMSSA algorithm applies to find the optimal flight path. Verified by simulation experiments, the improved multi-strategy salp swarm algorithm has a robust global search ability in the given simulation scenario, the convergence speed is obviously accelerated, and a better path is obtained compared with the given algorithm.

引用信息:Ruixiang Zhu, Taogang Hou, Yuxuan Guo, Xuan Pei, Junxin Zhu. Improved Multi-Strategy Salp Swarm Algorithm for UAV Path Planning. IEEE International Conference on Unmanned Systems (ICUS), 2023-10. DOI: 10.1109/icus58632.2023.10318445

论文链接:https://doi.org/10.1109/icus58632.2023.10318445