For my robotics course I have to implement the nonlinear continuous system function for a mobile robot that is equipped with a GPS receiver and an IMU consisting of an accelerometer and a gyroscope. For the prediction step with an EKF, I need to implement the kinematic model $\dot x = f(x,u)$ with given state vector $x = \begin{pmatrix}x\\y\\z\\v_x\\v_y\\v_z\\q_1\\q_2\\q_3\\q_4\end{pmatrix}$, input vector $u = \begin{pmatrix}a_x\\a_y\\a_z\\w_x\\w_y\\w_z\end{pmatrix}$, time $t$ and gravity $g$.
Afterwards, I need to calculate the matrix $A=\frac{\partial f}{\partial x} $ and $B=\frac{\partial f}{\partial u} $ where $A \in \mathbb{R}^{10x10}$ and $B \in \mathbb{R}^{10x6}$. These are used to calculate the system matrices $F$ and $H$ of the time discrete system (this is already given as Matlab code).
Unfortunately I didn't attend any control theory courses yet and it is my first semester in robotics, so I am pretty lost with this task. Maybe someone could give me a hint on how I have to approach this problem.
implement the [...] function
and then you say youneed to calculate the matrix A [...] and B
. Are you doing system identification? I (and I'm sure others) would love to help, but realistically the best person you're going to find for help on this is your TA, lab supervisor, professor, etc. $\endgroup$