Obtaining Linear State-Space Matrices (A,B,C,D) from UAV Toolbox Guidance Model for MPC/LQR Control Design

I need to design advanced controllers (MPC or LQR) for a UAV system using the Guidance Model block from UAV Toolbox. However, these control design methods require linear state-space matrices (A, B, C, D), but the Guidance Model represents a closed-loop reduced-order system that cannot be directly linearized using standard Simulink linearization tools.

Key Challenges :

1. Closed-Loop Nature: The Guidance Model already includes autopilot + airframe dynamics, making it a stabilized closed-loop system
2. Reduced-Order Model: Internal states don't directly correspond to physical states

Any guidance would be greatly appreciated!

Answer 1

[Jianxin Sun]

Hi, if you want to design controllers for the outter loop, such as position tracking controllers, you can still linearize the guidance model at different operating point. However, if you need to design controllers for the inner loop that is already part of guidance model, such as attitude controller for multirotor, you will need to build your model differently. For example, you can build your model using aerospace blockset following example in https://www.mathworks.com/help/aeroblks/quadcopter-modeling-based-on-parrot-minidrone.html

[boutegui]

When I type :

openProject('asbQuadcopter');

I get

Error using matlab.project.loadProject

The file or folder "asbQuadcopter" does not exist.

Error in openProject

[Jianxin Sun]

Have you ran the following command to get example files first?

openExample('aeroblks_quad/QuadcopterProjectExample')

[boutegui]

Thank you Sir