不确定随机系统LQG对偶控制

LQG Dual Control on Uncertain Stochastic Systems

针对不确定性模型为多胞型的随机系统进行了对偶控制的研究.首先,假定多胞型不确定参数存在于状态空间模型的系统矩阵、输入矩阵和量测矩阵中.其次,将此不确定性模型通过不确定参数的后验概率加权,得到其增广系统.然后,使用动态规划得到增广系统的次优控制律.最后,通过仿真验证了本文方法的可行性和有效性.

The problem of dual control is studied for the stochastic system of which the uncertainty model is polytopie. First, it is assumed that the polytope exists in the system matrix, input matrix and measure matrix of the state space model. Second, with the posterior probability weighting of the uncertain parameters, an augmented system is obtained for the uncertainty model. Then, the suboptimal control law is achieved using the dynamic programming principle. Finally, the feasibility and effectiveness of the presented approach are verified with simulation.