Ted Pavlic
The controller shown here was designed as part of a final project in ECE758 at The Ohio State University. For more information about ECE758, see:
http://www.ece.osu.edu/~passino/ee758.html
This controller is provided pitch and yaw position feedback from shaft encoders and must control the pitch and yaw position and velocity. Hence, a Luenberger observer is used to estimate the unknown states from a model of the 2DOF helicopter. Additionally, for lower steady-state error, the system model has been augmented to include integrators for both pitch and yaw. The linear feedback control gains for these six states are picked using LQR with an appropriate quadratic cost function (i.e., weighting matrices) for the helicopter system. The reference is generated by a joystick. In the video, students disturb the system (by pushing it) to verify that it will return to its set point.
The 2DOF helicopter is provided by Quanser academic. The controller is designed using Simulink (by Mathworks) and implemented on a dSPACE RTI1104 DSP.
ECE758: http://www.ece.osu.edu/~passino/ee758.html
Quanser academic: http://www.quanser.com/english/html/home/fs_homepage.html
Quanser 2DOF helicopter: http://www.quanser.com/english/html/products/fs_product_challenge.asp?lang_code=english&pcat_code=exp-spe&prod_code=S2-2dofheli&tmpl=1
Mathworks: http://www.mathworks.com/
dSpace: http://www.dspace.com/
LQR: http://en.wikipedia.org/wiki/Linear-quadratic_regulator
State observer: http://en.wikipedia.org/wiki/State_observer
Source
I should also emphasize that these were my students, and this was their project. This was not my project.
The control and estimation strategy is detailed in the long description of the video.
what's control method did you use in your project? #Ted Pavlic
Contact Quanser Academic for a quote on the 2DOF helicopter. You can find a link to the product page in the video description above.
If I would like to buy this Model from quanser (2 DOF helicopter)…How much does it cost?
x89codered89x — See the description provided with the video. In particular, this part:
"This controller is provided pitch and yaw position feedback from shaft encoders… Hence, a Luenberger observer is used to estimate the unknown states… Additionally, for lower steady-state error, the system model has been augmented to include integrators for both pitch and yaw. The linear feedback control gains for these six states are picked using LQR with an appropriate quadratic cost function…"