The Hamilton-Jacobi-Bellman equation with a gradient constraint (Q1114854)

From MaRDI portal
scientific article
Language Label Description Also known as
English
The Hamilton-Jacobi-Bellman equation with a gradient constraint
scientific article

    Statements

    The Hamilton-Jacobi-Bellman equation with a gradient constraint (English)
    0 references
    0 references
    1988
    0 references
    This paper deals with the existence and uniqueness of viscosity solutions for the Hamilton-Jacobi-Bellman equation \[ \max \{L^ 1u-f^ 1,...,L^ mu-f^ m,| Du| -g\}=0\quad in\quad \Omega;\quad u|_{\partial \Omega}=0. \] Here \(L^ p\), \(p=1,...,m\) are uniformly elliptic operators with smooth coefficients. For smooth \(f^ p\geq 0\), \(p=1,...,m\), and \(g\geq 0\), it is proved that there exists a viscosity solution which belongs to \(W^{2,\infty}_{loc}(\Omega)\cap W^{1,\infty}(\Omega)\) and it is unique in the class \(C^ 1(\Omega)\cap C({\bar \Omega})\) if \(g>0\). The solution is obtained as a limit of the system of approximate equations: \[ L^ p u^ p_{\epsilon}+\beta_{\epsilon}(| Du^ p_{\epsilon}|^ 2-g^ 2)+\gamma_{\epsilon}(u^ p_{\epsilon}- u_{\epsilon}^{p+1})=f^ p\quad in\quad \Omega, \] \[ u^ p_{\epsilon}|_{\partial \Omega}=0,\quad p=1,...,m,\quad u_{\epsilon}^{m+1}=u^ 1_{\epsilon}, \] where \(\beta_{\epsilon}\) and \(\gamma_{\epsilon}\) are standard penalty functions. The uniqueness proof is carried out by using this approximation again. The uniqueness of the solution in the class \(C(\Omega)\) is also considered and the author proves the affirmative results (1) for the case \(m=1\) of the above equation, (2) for the minimax equation \[ \min \{\max \{-\Delta u+u-f,u-\psi_ 1\},u-\psi_ 2\}=0\quad in\quad \Omega;\quad u|_{\partial \Omega}=0. \]
    0 references
    existence
    0 references
    uniqueness
    0 references
    viscosity solutions
    0 references
    Hamilton-Jacobi-Bellman equation
    0 references
    uniformly elliptic
    0 references
    smooth coefficients
    0 references
    limit
    0 references
    approximate equations
    0 references
    standard penalty functions
    0 references
    minimax equation
    0 references

    Identifiers

    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references