mvirgo
commented
5 years ago I would not say this is extraneous. It's always important to know about your data prior to doing anything with it, so it's still trying to encourage everyone to take a look at it. Additionally, it's useful to know what else is actually happening within main.cpp, and otherwise some of the code around measurement package could be fairly confusing or ambiguous.
We can investigate whether it makes sense to refactor it to be able to use either source; I think that's a pretty good idea here.
Am I missing something? The "obj_pose-laser-radar-synthetic-input.txt" appears to be unrefenced from main.cpp because now the measurements data source is provided from the sim via callbacks. If this is true then this lesson should be removed: It is confusing mis-information.
On the other hand I think the old architecture is a better design for dev/debug of the Kalman/Filter: Trying to dev/debug code that is only invoked in a callback from another process adds a layer of unneeded complexity for getting the EKF code to work. I think the code should be refactored to permit selection of either data source.
And specifically supplying a long anonymous callback function definition to
h.onMessage(...is a obscuring javascript like behaviour. It's difficult to comprehend and it's impossible to test against. An