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

this paper deals with the infinite-horizon optimal control problem, by a sutiable change of variable, we transfrom 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 construction an error fumction, we define an unconstrained minmization problem. in the optimization problem, we use trial solutions for the state, costate and control functions where these trial solution are constructed by using wto layered perceptron neural network, an application is introduced to explain our main results.

Keywords:

infinite horizon problems , pontryagin minimum principle , Optimal control problem , neural networks