In this paper, neural networks are used to approximately solve the finite-horizon constrained input H-infinity state feedback control problem. The method is based on solving a related Hamilton-Jacobi-Isaacs equation o...In this paper, neural networks are used to approximately solve the finite-horizon constrained input H-infinity state feedback control problem. The method is based on solving a related Hamilton-Jacobi-Isaacs equation of the corresponding finite-horizon zero-sum game. The game value function is approximated by a neural network with time- varying weights. It is shown that the neural network approximation converges uniformly to the game-value function and the resulting almost optimal constrained feedback controller provides closed-loop stability and bounded L2 gain. The result is an almost optimal H-infinity feedback controller with time-varying coefficients that is solved a priori off-line. The effectiveness of the method is shown on the Rotational/Translational Actuator benchmark nonlinear control problem.展开更多
A new control mode is proposed for a networked control system whose network-induced delay is longer than a sampling period. A time-division algorithm is presented to implement the control and for the mathematical mode...A new control mode is proposed for a networked control system whose network-induced delay is longer than a sampling period. A time-division algorithm is presented to implement the control and for the mathematical modeling of such networked control system. The infinite horizon controller is designed, which renders the networked control system mean square exponentially stable.Simulation results show the validity of the proposed theory.展开更多
针对电离层虚高观测误差使得超视距目标定位精度受到一定限制的问题,提出了一种电离层虚高误差影响下基于神经网络的单站定位误差校正方法。在存在电离层虚高误差的单站定位场景中,利用神经网络对多重信号分类(Multiple Signal Classifi...针对电离层虚高观测误差使得超视距目标定位精度受到一定限制的问题,提出了一种电离层虚高误差影响下基于神经网络的单站定位误差校正方法。在存在电离层虚高误差的单站定位场景中,利用神经网络对多重信号分类(Multiple Signal Classification,MUSIC)算法的位置估计结果进行修正。仿真实验表明,在信噪比为0 dB的情况下,30 km的电离层虚高误差引起的定位误差能从111 km降到10 km以内。展开更多
构造一种线性差分式Hop fie ld网络(LDHNN),其稳定状态可使能量函数达到唯一极小值.利用该网络稳定性与其能量函数收敛特性的关系,提出了基于LDHNN的移动域控制方法.LDHNN的理论设计表明,网络的稳态输出即为移动域LQ控制问题的解.当系...构造一种线性差分式Hop fie ld网络(LDHNN),其稳定状态可使能量函数达到唯一极小值.利用该网络稳定性与其能量函数收敛特性的关系,提出了基于LDHNN的移动域控制方法.LDHNN的理论设计表明,网络的稳态输出即为移动域LQ控制问题的解.当系统满足一定条件时,基于LDHNN的移动域LQ控制能保证闭环最优控制系统的渐近稳定性.数字仿真取得了与理论分析一致的实验结果.展开更多
基金This work was supported by the National Science Foundation (ECS-0501451)Army Research Office (W91NF-05-1-0314).
文摘In this paper, neural networks are used to approximately solve the finite-horizon constrained input H-infinity state feedback control problem. The method is based on solving a related Hamilton-Jacobi-Isaacs equation of the corresponding finite-horizon zero-sum game. The game value function is approximated by a neural network with time- varying weights. It is shown that the neural network approximation converges uniformly to the game-value function and the resulting almost optimal constrained feedback controller provides closed-loop stability and bounded L2 gain. The result is an almost optimal H-infinity feedback controller with time-varying coefficients that is solved a priori off-line. The effectiveness of the method is shown on the Rotational/Translational Actuator benchmark nonlinear control problem.
文摘A new control mode is proposed for a networked control system whose network-induced delay is longer than a sampling period. A time-division algorithm is presented to implement the control and for the mathematical modeling of such networked control system. The infinite horizon controller is designed, which renders the networked control system mean square exponentially stable.Simulation results show the validity of the proposed theory.
文摘构造一种线性差分式Hop fie ld网络(LDHNN),其稳定状态可使能量函数达到唯一极小值.利用该网络稳定性与其能量函数收敛特性的关系,提出了基于LDHNN的移动域控制方法.LDHNN的理论设计表明,网络的稳态输出即为移动域LQ控制问题的解.当系统满足一定条件时,基于LDHNN的移动域LQ控制能保证闭环最优控制系统的渐近稳定性.数字仿真取得了与理论分析一致的实验结果.
