Nonlinear adaptive neural control with limited actuator authority

OData support
Supervisor:
Dr. Harmati István
Department of Control Engineering and Information Technology

In this thesis I extend the method used in (Georgia Institute of Technology Phd. Thesis, Limited authority adaptive flight control, 2000, Eric N. Johnson) to a more general system class. In the thesis I will point out the differences between the resulting equations. In order to reduce the number of equations in the thesis I left out some calculations. After this extension I separately considered the case, where the plant is not locally controllable.

I checked the results by simulation in the Matlab Simulink environment, and the results are expected, based on the calculations, and the control loop is stable among very wide limits.

I also started to build the components to connect to hardware. Testing on the real hardware is not yet possible, because:

1. Currently the controller is too slow.

2. Parallel run of the image processing and control is not solved yet.

These problems will be addressed in my final thesis, with the goal to create a controller for the real hardware, achieving acceptable performance.

Downloads

Please sign in to download the files of this thesis.