计算机工程与应用 ›› 2020, Vol. 56 ›› Issue (4): 209-213.DOI: 10.3778/j.issn.1002-8331.1810-0397

• 工程与应用 • 上一篇    下一篇

RS-GA神经网络无人机受风情况估计

张博,贾华宇,马珺   

  1. 1.太原理工大学 物理与光电工程学院,太原 030024
    2.太原理工大学 电气与动力工程学院,太原 030024
  • 出版日期:2020-02-15 发布日期:2020-03-06

Estimation of Air Condition for Unmanned Aerial Vehicle Based on RS-GA Neural Network

ZHANG Bo, JIA Huayu, MA Jun   

  1. 1.College of Physics and Optoelectronics, Taiyuan University of Technology, Taiyuan 030024, China
    2.College of Electrical and Power Engineering, Taiyuan University of Technology, Taiyuan 030024, China
  • Online:2020-02-15 Published:2020-03-06

摘要:

针对现有无人机(Unmanned Aerial Vehicle,UAV)风场估计方法中存在的计算复杂、需额外搭载传感器等问题,提出基于粗糙集遗传神经网络的无人机受风状态估计方法。该方法利用粗糙集分析方法对无人机上采集的姿态信息数据集进行约简;利用遗传算法全局搜索能力强的特点优化神经网络的初始权值;用简化的无人机数据集训练神经网络即得到所需神经网络风场估计模型。仿真结果表明,该方法具有较高的识别率以及较短的训练时间,证明了其在无人机风场估计上应用的有效性。

关键词: 无人机(UAV), 受风状态估计, 粗糙集, 约简, 遗传算法, 神经网络

Abstract:

Aiming at the problems of complex calculation and additional sensors in the existing Unmanned Aerial Vehicle(UAV) wind field estimation method, a wind estimation method based on rough set genetic neural network is proposed. The method uses the rough set analysis method to reduce the attitude information data set collected on the UAV, optimizes the initial weight of the neural network by using the global search ability of the genetic algorithm, and trains the neural network with the simplified UAV data set. The required neural network wind field estimation model is obtained. The simulation results show that the proposed method has higher recognition rate and shorter training time, which proves its effectiveness in UAV wind field estimation.

Key words: Unmanned Aerial Vehicle(UAV), wind estimation, rough set, reduction, genetic algorithm, neural network