Linear Quadratic Optimal Control problem.pdf

Chapter 6 LINEAR QUADRATIC OPTIMAL CONTROL In this chapter, we study a different control design methodology, one which is based on optimization. Control design objectives are formulated in terms of a cost criterion. The optimal control law is the one which minimizes the cost criterion. One of the most remarkable results in linear control theory and design is that if the cost criterion is quadratic, and the optimization is over an infinite horizon, the resulting optimal control law has many nice properties, including that of closed loop stability. Furthermore, these results are intimately connected to system theoretic properties of stabilizability and detectability. 6.1 Quadratic Forms Before we state the optimal control problem, we review briefly the concept of quadratic forms. Let S be an n × n symmetric matrix, i.e., S = ST . For x ∈ Rn , the scalar-valued function xT Sx is called a quadratic form associated with S . It is known from linear algebra that if S is symmetric, all its eigenvalues are real, and it is diagonalizable by an orthogonal matrix M , meaning that M −1 = M T . If every quadratic form associated with S , xT Sx ≥ 0, we say S is positive semidefinite and write S ≥ 0. If for any x = 0, xT Sx 0, we say S is positive definite and write S 0. Example: The matrix 1 0 S1 = 0 0 is positive semidefinite, while the matrix 1 0 S2 = 0 4 is positive definite. It can be shown that S ≥ 0 if and only if all its e