Closed Nicogene closed 1 year ago
fyi @GiulioRomualdi @S-Dafarra @gabrielenava this may affect any code that is readying directly from FT sensor ports as a vector.
One thing that we may took the occasion to do know that we do a breaking change is that we could group in a single wrapper all the FT sensors of the leg, for consistency with what we do for other wrappers, such as controlboard.
So, it would be much easier to document which FT sensors are available, i.e. :
For each body part, the available FT and temperature sensors are exposed with a
multipleanalogsensorsserver
with a prefix/<robotName>/<bodyPart>
. If nomultipleanalogsensorsserver
with a prefix/<robotName>/<bodyPart>
is present, no FT or temperature sensors are avalable on that part (for examplehead
ortorso
as of iCub v2.9).
Instead, in the current situation the ports on which the FT sensors are published are essentially undocumented, complicating the life for users that want to access them. Furthemore, it may be a bit confusing that a multipleanalogsensorsserver
that only contains FT sensors and temperature sensors (and no IMU, for example) is published on ports with prefix /<robotName>/<bodyPart>
, while for example IMUs (except for xsens's one, but that is another story) are published on /<robotName>/<bodyPart>/inertials
.
I know that in general it would be better to do this changes incrementally, but in this case doing several breaking changes may be problematic, so a breaking change in this sense it is a useful opportunity.
rebased on top of the latest devel
.
cc @Nicogene
@traversaro in 417254c
(#482) I changed /<robotName>/<bodyPart>
to /<robotName>/<bodyPart>/FT
, we can think about if we want to introduce also this change or I can revert it.
After a T2T alignment with @traversaro and @pattacini we decided to proceed merging this PR(it ha been tested on iCubGenova11
) and then address
One thing that we may took the occasion to do know that we do a breaking change is that we could group in a single wrapper all the FT sensors of the leg, for consistency with what we do for other wrappers, such as controlboard.
in a second stint; https://github.com/robotology/robots-configuration/issues/485
This has been done both for uniforming with the simulated robots: https://github.com/robotology/icub-models-generator/pull/231 and for the deprecation of
analogServer
It fixes #470
To be tested on a robot