Design & Implementation of State Estimation Based Optimal Controller Model Using MATLAB/SIMULINK
Muhammad Junaid Rabbani1, Asim-ur-Rehman khan2
1Muhammad Junaid Rabbani, Electrical Engineering, National University of Computer & Emerging Sciences-FAST, Karachi, Pakistan.
2Asim-ur-Rehman khan, Electrical Engineering, National University of Computer & Emerging Sciences-FAST, Karachi, Pakistan.
Manuscript received on September 11, 2013. | Revised Manuscript received on September 15, 2013. | Manuscript published on September 25, 2013. | PP: 64-69 | Volume-1, Issue-11, September 2013. | Retrieval Number: K04780911113/2013©BEIESP
Open Access | Ethics and Policies | Cite | Mendeley
© The Authors. Published By: Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: This paper proposes model based technique to control the horizontal position of a helicopter. The main constraint in the controller design is that not many states are measurable, and that the available sensor information is highly corrupted by noise. Here, the change in the declining angle of rotor is used to steer the helicopter in a straight line. The design is constrained to keep other parameters within the specified limits. The controller design is based on a combination of Kalman filter observer along with optimal linear quadratic Gaussian (LQG) controller. The design is implemented in two steps. First, Kalman filter is used to design an observer that estimates two desired states of a helicopter: rotator angle and horizontal position. Second, state feedback controller gain is estimated using the linear quadratic criterion function. The state controller enhances the regulation performance, while minimizing cost of control effort. Simulation results prove the credibility of Kalman filter observer by comparing the estimated states such as position and angle with the model output. In addition, the performance of LQG controller is examined by incorporating servo control mode that reduces the effort to compute error between reference and measured position.
Keywords: Helicopter system, Kalman filter observer, linear quadratic gaussian controller, state space model