Zhong Jun, Gou Xin, Shu Qin, Liu Xing, Zeng Qi
School of Electrical Engineering, Sichuan University, Chengdu 610000, China.
Sensors (Basel). 2021 Feb 2;21(3):997. doi: 10.3390/s21030997.
Foreign object debris (FOD) on airport runways can cause serious accidents and huge economic losses. FOD detection systems based on millimeter-wave (MMW) radar sensors have the advantages of higher range resolution and lower power consumption. However, it is difficult for traditional FOD detection methods to detect and distinguish weak signals of targets from strong ground clutter. To solve this problem, this paper proposes a new FOD detection approach based on optimized variational mode decomposition (VMD) and support vector data description (SVDD). This approach utilizes SVDD as a classifier to distinguish FOD signals from clutter signals. More importantly, the VMD optimized by whale optimization algorithm (WOA) is used to improve the accuracy and stability of the classifier. The results from both the simulation and field case show the excellent FOD detection performance of the proposed VMD-SVDD method.
机场跑道上的外来物碎片（FOD）会导致严重事故和巨大经济损失。基于毫米波（MMW）雷达传感器的FOD检测系统具有更高的距离分辨率和更低的功耗等优点。然而，传统的FOD检测方法很难从强地面杂波中检测和区分目标的微弱信号。为了解决这个问题，本文提出了一种基于优化变分模态分解（VMD）和支持向量数据描述（SVDD）的新型FOD检测方法。该方法利用SVDD作为分类器来区分FOD信号和杂波信号。更重要的是，采用鲸鱼优化算法（WOA）优化的VMD来提高分类器的准确性和稳定性。仿真和现场案例的结果都表明了所提出的VMD-SVDD方法具有出色的FOD检测性能。
