During the design of control strategies, the inclusion of constraints to the control signal applicable to the actuators may lead to several challenges in fulfilling performance metrics related to the control law. And nevertheless, the proper control strategies are not always used despite the presence of constraints or the constraints are not always properly included in the control problem. In this paper, different control strategies are compared for a set point tracking problem in a narrowed constraints scenario in order to quantify the real performance degradation when constraints are not properly treated. This scenario minds the actuators bounds and mechanical constraints, the system being a Quanser 2 DoF Helicopter. A model of the helicopter is first developed and its constraints are shown explicitly. Then, several control strategies are designed from closed-loop performance specifications in the time domain: PID, State-Feedback, LQR and two variants of MPC (MPC and NMPC), for set point tracking. Each of them is compared in simulation as constraints get tighter and several closed-loop time-domain performance metrics are measured. The inclusion of the constraints prevents some of the classical control strategies to fulfil the performance metrics, whereas advanced modern control techniques, such as MPC, are able to fulfil them.
Tópico:
Adaptive Control of Nonlinear Systems
Citaciones:
0
Citaciones por año:
No hay datos de citaciones disponibles
Altmétricas:
0
Información de la Fuente:
Fuente2017 IEEE 3rd Colombian Conference on Automatic Control (CCAC)