Based on input-output data, state space models are directly identified by the SMI methods combined with the system theory, linear algebra and statistics. 这类方法综合了系统理论，线性代数和统计学三方面的思想，其特点是直接由输入输出数据辨识系统的状态空间模型，因而非常适合多变量系统辨识。
In the self-developed multivariable system design platform for a parameter uncertainty robust industrial object system design shows its effectiveness and practicality. 在自行研制的多变量系统设计平台中对一个参数不确定性工业对象的鲁棒系统进行设计，结果表明了该算法的有效性和实用性。
