针对仿射非线性系统的最优跟随控制问题,提出了一种基于最小二乘支持向量机(least squares support vector machine,LS-SVM)的数据驱动方法.通过非线性系统已知信息和期望轨迹的离散数据构建LS-SVM模型,获得最优跟随轨线的近似解并求得...针对仿射非线性系统的最优跟随控制问题,提出了一种基于最小二乘支持向量机(least squares support vector machine,LS-SVM)的数据驱动方法.通过非线性系统已知信息和期望轨迹的离散数据构建LS-SVM模型,获得最优跟随轨线的近似解并求得最优跟随控制器,使系统达到期望的动态性能.数值算例仿真证实,该方法具有优化和学习能力,能够实现在较小误差范围内对期望轨迹的准确跟踪.展开更多
This paper discusses a problem of optimal tracking for a linear control system driven by fractional Brownian motion.An equation is obtained for the linear Markov feedback control.The existence and uniqueness of the so...This paper discusses a problem of optimal tracking for a linear control system driven by fractional Brownian motion.An equation is obtained for the linear Markov feedback control.The existence and uniqueness of the solution to the equation are also studied.展开更多
文摘针对仿射非线性系统的最优跟随控制问题,提出了一种基于最小二乘支持向量机(least squares support vector machine,LS-SVM)的数据驱动方法.通过非线性系统已知信息和期望轨迹的离散数据构建LS-SVM模型,获得最优跟随轨线的近似解并求得最优跟随控制器,使系统达到期望的动态性能.数值算例仿真证实,该方法具有优化和学习能力,能够实现在较小误差范围内对期望轨迹的准确跟踪.
基金partially supported by a grant from the Simons Foundation #209206
文摘This paper discusses a problem of optimal tracking for a linear control system driven by fractional Brownian motion.An equation is obtained for the linear Markov feedback control.The existence and uniqueness of the solution to the equation are also studied.
