In doing the exercises, then the project and looking at how my own robots have configured the parameters for the navigation stack it seems there are many different ways of doing this. In the exercises
my_move_base_params.yaml has the move base params in it along with the Navfn, TrajectoryPlannerROS and DWAPlannerROS all in the same file ex.4.6 and 4.8 for example. The project has separate files and my robots are different again. In researching this I found even more variations. So how do you ever arrive at a given solution. More importantly how do you know the best way to organize them for a particular rorbot. It almost seems like applying our knowledge in a trial and error fashion until we find a configuration that works. Unless I’m missing something. So is there a recommended way to organize these parameters and files or a least a good place to start ?
Hello @hskramer ,
Yes, that’s right. In the end, they are all parameters that are loaded to the ROS parameter server, so it doesn’t really matter HOW you load them. You could even load them all in one file!
I’d say that, in this case, there’s not really a correct answer. The best practice is to load them as it is better (or more clear) for you. My suggestion, though, is to divide them more or less as we indicate in the course:
-
One file for the move_base parameters and the planners. These files can also be separated if you prefer so (you can also separate the global planners from the local planners if you prefer it like that).
-
One file for the common costmap parameters.
Best,
-
One file for the global costmap parameters.
-
One file for the local costmap parameters.