Quantcast

Simplifying use of paparazzi software for academics with limited or no paparazzi experience

classic Classic list List threaded Threaded
1 message Options
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Simplifying use of paparazzi software for academics with limited or no paparazzi experience

Ali Jamei
This post has NOT been accepted by the mailing list yet.
Dear paparazzi users,

I would like to propose to work on a new concept for paparazzi software development that would change the way it is used among users with limited paparazzi experience, but first let me introduce myself.

My name is Ali Jamei and I'm a MSc. student of Aerospace engineering, flight mechanics. I have been using paparazzi software for almost 3 years on different airframes, from hand launched aircraft to Ar.Drone 2 quadrotors. I have developed and tested different control and guidance algorithms for different air-frames in paparazzi software, for the purpose of research and I have gained a lot of experience in developing and debugging new codes in paparazzi. I have met many of developers including Gautier Hattenberger with ENAC team and Christophe De Wagter with Tu-delft team, during IMAV 2014 and 2015 competitions, in person.

I would like to work on paparazzi software as my MSc. thesis with the goal of making this nice project better understandable and used for users with no/limited experience and also extending the capabilities of this project to a new level. I would like to do this as follows:

1) preparing a new documentation of workflow of the whole project and how it works generally so that the new user not be confused.
2) preparing documentation for the current control loop codes and explaining them(including the ones that aren't in the wiki now).
3) the most important step, developing a new interface to connect paparazzi and MATLAB. Many academics around the world like MATLAB better than anything when trying to design and simulate new concepts, i.e. control loops or navigation filters. So connecting MATLAB to paparazzi in way that eases C programming in paparazzi for the user would be a great idea for fast algorithm prototyping and tuning rather than manually writing C codes in paparazzi. I'm talking about some kind of code generation in MATLAB that is compatible with paparazzi software practically.  
4) simulating current control loops in MATLAB as an example of No.3
5) providing documentation for MATLAB-paparazzi interface and committing the codes to github for everyone to use.
6) designing a new attitude controller for rotorcrafts in MATLAB using the developed interface and testing it in the real world using paparazzi , as an example of using this interface.

This is my game plan so far on how to make this happen, But before I start this endeavor, I would like you to help me with your ideas. Any ideas on this matter or the steps or the whole thing would be appreciated. Needless to say that doing this would take a lot of time and effort but I should note that I'm not alone in this and I intend to this as my MSc. thesis.

Any thoughts???
Loading...