A New RBF Neural Network with GA-based Fuzzy C-Means Clustering Algorithm for SINS Fault Diagnosis

摘要：

In this paper, a new radial basis function (RBF) neural network with fuzzy c-means clustering algorithm based on genetic algorithm (GA) is proposed for the fault diagnosis of gyroscopes and accelerometers of strapdown inertial navigation system (SINS). The fuzzy c-means algorithm (FCM) tends to fall into the local optimum. The fuzzy c-means clustering algorithm combined with GA (FGA) obtains the global optimal cluster centers. FGA is used to provide the optimal cluster centers for RBF neural network, and a second order learning algorithm is used to train the parameters and weights of RBF neural network. Experimental results show that the proposed RBF neural network with FGA quickly converges and effectively improves the diagnostic accuracy rate of SINS fault diagnosis.

关键词： Radial basis function neural network Fault diagnosis Strapdown inertial navigation system Fuzzy c-means clustering algorithm Genetic algorithm

作者: Zhide Liu Jiabin Chen Chunlei Song

作者单位: School of Automation, Beijing Institute of Technology, Beijing 100081, China

会议类型: 国际会议

会议名称: 2009年中国控制与决策会议(2009 Chinese Control and Decision Conference)

会议地点: 广西桂林

会议语种:英文

页码: 208-211

在线出版日期: 2009-06-17（万方平台首次上网日期，不代表论文的发表时间）