-
Róger Mateus Sehnem
Programa de Pós Graduação em Engenharia Elétrica (PPGEE), Universidade Federal do Rio Grande do Sul, RS
-
Alexandre Sanfelici Bazanella
Programa de Pós Graduação em Engenharia Elétrica (PPGEE), Universidade Federal do Rio Grande do Sul, RS
Keywords:
Optimal Closed Loop Control, DNN, Nonlinear Dynamics, Inverted Pendulum, Nonlinear Control
Abstract
In this work, the design of an optimal closed-loop control of an inverted pendulum is performed using a deep neural network (DNN). A training database is created by solving several optimal control problems, resulting in an optimal open loop control, using the MATLAB platform. The DNN is then trained in this database and, after, used to calculate the optimal control given the plant states, effectively closing the loop. Simulations are performed and presented to demonstrate the optimal controller performance when applied to the system.
