Gongxing Wu

Learn More
Research on sensor fault diagnosis of underwater robots (URs) is undertaken to improve its whole system reliability. Based on the analysis of URspsila three kinds of sensor failures, fault diagnosis methods corresponding to these failures are presented. The basic principles of wavelet transform and linear smoothing are expressed, signal singularity analysis(More)
Research on the design of the basic motion control system of the water-jet propelled Unmanned Surface Vehicle (USV) is undertaken. The architecture of the USV'S embedded motion control system is presented from the view point of the hardware and software. The characteristic of the equipment configuration of the USV'S motion control actuators was analysed,(More)
According to the characteristic of unmanned surface vehicle(USV) based on water-jet propulsion, four-degree of freedom maneuvering motion equation was established. External forces acting on USV were divided into gravity, buoyancy, inertial hydrodynamic force, viscous hydrodynamic force, dynamic lift force and water-jet force. Momentum theorem was used to(More)
Researches are undertaken to improve the control effect of Underwater Robots. A capacitor plate model controller (CPMC) is designed for a certain type of Underwater Robot. A new control method: generalized S-plane controller (GSPC) is put forward based on the analysis of two new control methods, sigmoid S-plane control (SSPC) and capacitor plate model(More)
Research on thruster fault diagnosis of Underwater Robots (URs) is undertaken to improve its whole system reliability. Based on the BP neural network, a recurrent neural network (RNN) is presented and the network training algorithm is deduced. The RNN is trained by voyage head and yaw turning experiments, and the well trained network is applied to model for(More)
Study of thruster fault diagnosis of Underwater Robots (URs) is undertaken to improve its whole system reliability. Based on the BP neural network, an improved recurrent neural network (RNN) is proposed and the network training algorithm is deduced. The RNN is trained by voyage head and yaw turning experiments, and the well trained network is applied to(More)
  • 1