Solving infinite-horizon optimal control problems with a feed forward neural network scheme

This paper deals with the infinite-horizon optimal problem. By a suitable change of variable, we transform the problem to a finite-horizon optimal control problem (FHOC) problem. Then according to the Pontryagin minimum principle (PMP) for optimal control problems and by constructing an error function, we define an unconstrained minimization problem. In the optimization problem. We use trial solutions for the state, costate functions where these trial solution are constructed by using two-layered neural network. An application is introduced to explain our main results.