Closed traversaro closed 4 days ago
To be honest, I am wondering how we were removing the FT from the estimation with this bug. Perhaps in all other cases were also either avoiding to pass the upper_leg FTs devices to the attach list, or we removed the FT sensors from the devices opened by the yarprobotinterface?
Both 😅
I was looking at this piece of code https://github.com/robotology/whole-body-estimators/blob/1fc1d54c9db9183f7a6e4a2f537ea8bef885230c/devices/wholeBodyDynamics/WholeBodyDynamicsDevice.cpp#L1826-L1836
Maybe it should throw an error if the name is not found?
Maybe it should throw an error if the name is not found?
I am preparing a different PR to drop the support for reading FT sensor data from IAnalogSensor interfaces, I think it is more appropriate to keep that change there.
Maybe it should throw an error if the name is not found?
I am preparing a different PR to drop the support for reading FT sensor data from IAnalogSensor interfaces, I think it is more appropriate to keep that change there.
Actually that was already removed in https://github.com/robotology/whole-body-estimators/pull/181 . However I indeed preparing a different cleanup PR.
This is a bug found by @LoreMoretti , basically he was trying to remote just the upper_leg FTs from ergocub from the estimation, but the wbd launch was failing with:
However, the check does not make a lot of sense.
nrMASFTSensors
is the total number of FT sensors in all the devices attached to wbd (for ergocub 8), whileremappedMASInterfaces.ftMultiSensors->getNrOfSixAxisForceTorqueSensors()
andftMultipleAnalogSensorNames.size()
are the actual number of sensors considered (removing theupper_leg
fts as @LoreMoretti was doing, 6).To be honest, I am wondering how we were removing the FT from the estimation with this bug. Perhaps in all other cases were also either avoiding to pass the upper_leg FTs devices to the attach list, or we removed the FT sensors from the devices opened by the yarprobotinterface?