针对无人机舵面电动加载系统具有非线性及多余力矩的特点,提出了一种自适应CMAC(Cerebellar Model Articulation Controller)神经网络与自适应神经元控制器并联构成复合控制结构.该控制策略以系统的指令输入和实际输出作为CMAC的激励信...针对无人机舵面电动加载系统具有非线性及多余力矩的特点,提出了一种自适应CMAC(Cerebellar Model Articulation Controller)神经网络与自适应神经元控制器并联构成复合控制结构.该控制策略以系统的指令输入和实际输出作为CMAC的激励信号,以系统的当前控制误差作为CMAC的训练信号.提出了利用误差在线自适应调整学习率的方法,消除了常规前馈型CMAC的过学习和不稳定现象.建立了无人机舵面电动加载系统的数学模型,给出了具体的控制结构和算法.仿真结果表明:该方法有效抑制了加载系统的多余力矩,增强了系统的稳定性,明显改善了舵面电动加载系统的动态性能.展开更多
A terminal sliding mode fuzzy control based on multiple sliding surfaces was proposed for ship course tracking steering, which takes account of rudder characteristics and parameter uncertainty. In order to solve the p...A terminal sliding mode fuzzy control based on multiple sliding surfaces was proposed for ship course tracking steering, which takes account of rudder characteristics and parameter uncertainty. In order to solve the problem, the controller was designed by employing the universal approximation property of fuzzy logic system, the advantage of Nussbaum function, and using multiple sliding mode control algorithm based on the recursive technique. In the last step of designing, a nonsingular terminal sliding mode was utilized to drive the last state of the system to converge in a finite period of time, and high-order sliding mode control law was designed to eliminate the chattering and make the system robust. The simulation results showed that the controller designed here could track a desired course fast and accurately. It also exhibited strong robustness peculiarly to system, and had better adaptive ability than traditional PID control algorithms.展开更多
文摘针对无人机舵面电动加载系统具有非线性及多余力矩的特点,提出了一种自适应CMAC(Cerebellar Model Articulation Controller)神经网络与自适应神经元控制器并联构成复合控制结构.该控制策略以系统的指令输入和实际输出作为CMAC的激励信号,以系统的当前控制误差作为CMAC的训练信号.提出了利用误差在线自适应调整学习率的方法,消除了常规前馈型CMAC的过学习和不稳定现象.建立了无人机舵面电动加载系统的数学模型,给出了具体的控制结构和算法.仿真结果表明:该方法有效抑制了加载系统的多余力矩,增强了系统的稳定性,明显改善了舵面电动加载系统的动态性能.
基金the National Natural Science Foundation ofChina (60974136)
文摘A terminal sliding mode fuzzy control based on multiple sliding surfaces was proposed for ship course tracking steering, which takes account of rudder characteristics and parameter uncertainty. In order to solve the problem, the controller was designed by employing the universal approximation property of fuzzy logic system, the advantage of Nussbaum function, and using multiple sliding mode control algorithm based on the recursive technique. In the last step of designing, a nonsingular terminal sliding mode was utilized to drive the last state of the system to converge in a finite period of time, and high-order sliding mode control law was designed to eliminate the chattering and make the system robust. The simulation results showed that the controller designed here could track a desired course fast and accurately. It also exhibited strong robustness peculiarly to system, and had better adaptive ability than traditional PID control algorithms.
