No label defined (Q4514461)

The authors study the Dirichlet type boundary value problem \[ F(x,u,Du,D^2u)=0, \quad x\in \Omega,\tag{1} \] \[ u=g \quad or \quad F(x,u,Du,D^2u)=0,\quad x\in \partial\Omega, \] where \(\Omega\subset \mathbb{R}^N\) is an open set, \(F\) is a fully nonlinear nondegenerated elliptic operator increasing in the variable \(u\), and \(g\) is continuous on \(\partial\Omega\). They consider the viscosity and the extended solution of type (1) problems and choose (under some assumption about their behaviour) the maximal subsolution \(u\) as a correct generalized solution to (1) which allows it to be discontinuous. The solution defined in this way is called the e-solution or the Perron-Wiener solution and the approximation theorems are given to construct it, in particular by penalization. The results are applied to study the Hamilton-Bellman equations of time optimal stochastic control and pursuit-evasion games.