Solver design especially in a computer simulation is an important matter to emulate any physical dynamical system. In order to understand the underlying mechanism of this thing, there should be a model needs to be developed in terms of differential equations and matrices equations.
In this study, a system dynamic library is developed in order to simulate any kind of system with respect to the given time parameters and system models.
Also, the main course of this GitHub repository includes the bare metal Kalman filter design in Matlab.
There are four folders in this repository:
- Learning : standard matlab kalman filter based on forward difference solver in linear domain (MCK system)
- Nonlinear : nonlinear system example with van der pol equation, jacobian usage and real time update
- simulink : linear and nonlinear system applications given in the folder of nonliner/learning, MCK and Van der pol equation are performed!
- unmodeledKalman : this is the simplest real time low pass filter approach on the data coming from the unknown dynamic
There are five main usages of this library:
- filtering the any signal on embedded systems
- system identification and estimation
- simulation of the modeled system
- obtaining time response characteristics of the system
- creating solver design to be implemented on embedded systems
Which kinds of projects you can utilize the basis of these codes?
- Engine design (physic motors)
- VR/AR application
- Embedded software simulation
- Estimation and Control of Physical Systems
- Real time engine application
How can you use the library?
- Suppose that you have a system like xdot = Ax + Bu with the parameter of A matrice (state matrice) and B matrice (input matrice). x is a vector representing states and u is an input vector.
- You can set the whole parameters related to discrete sampling period and you wil get time response of system in Matlab script and simulink versions