[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[Paparazzi-devel] ins subsystem
From: |
Gautier Hattenberger |
Subject: |
[Paparazzi-devel] ins subsystem |
Date: |
Thu, 12 Jul 2012 13:53:34 +0200 |
User-agent: |
Mozilla/5.0 (X11; Linux x86_64; rv:10.0.4) Gecko/20120510 Icedove/10.0.4 |
Hi everyone,
For people working with the master branch (development branch), after
commit
https://github.com/paparazzi/paparazzi/commit/6b875f40dff3a61873e343cb6472b3cc0e2b357c,
a separated ins subsystem has been introduced for _rotorcrafts_.
This will allow to handle more easily several ins implementations.
What you need to do in your airframe file is to add this subsystem:
<subsystem name="ins"/>
With this configuration, you will have the vertical filter (float
version) running with accelerometers and barometer, and the raw gps
position and velocities for the horizontal part.
All airframe files in the git repository have been updated.
You can activate an horizontal filter based on gps and accelerometers by
adding the line:
<subsystem name="ins" type="hff"/>
Please note that this code, while working, doesn't improve that much the
estimation and should be completely re-written (anyone with an urgent
need to do some kalman filtering ?).
You will also notice an "extended" ins subsystem, which is an attempt to
properly use an other altitude information, like a sonar, to correct the
barometer drift. This code is working as long as the flat earth
assumption is correct (flying over obstacles will produce altitude step
with a sonar) and that the input signal is not going crazy (which can be
the case with ultrasonic sensors close to motors and propellers).
The road-map for this ins subsystem is to split the existing one into
several, each corresponding to a given configuration (one of the
vertical/horizontal filter or none). This should improve the readability
of the code.
Do not hesitate to report issues.
Good flights to everyone
Gautier