[cob_hardware_config] further simplify configuration structure (modules and components)

[fmessmer]

The structure already got much simpler but there still is some potential.

I'm creating this issue for discussion following

• https://github.com/ipa320/cob_common/pull/175#issuecomment-154128294
• [WIP] add sensorring sensors within cob_hardware_config #359

This has minor priority for now...

@fmessmer FYI

[fmessmer]

After cleaning up unsupported robots #588, we will start introducing a more generic robot-/module-/component-base structure for all URDF, launch and config files... as discussed with @ipa-fmw @ipa-nhg @ipa-mdl

This will allow further de-duplication of code, more consistency between robots, better harmonization of HW/Sim-behavior as well as easier future maintenance and up-scaling...

[fmessmer]

#622 is a significant step towards a simplified configuration structure

I will apply a similar concept for the other "unifiable" components, e.g. torso, head, sensorring, arm(s),...

[fmessmer]

Also, we would need to adjust the structure to the robot names we are using now to refer to a (physically existing) robot composition, i.e. cob4-paul-ingolstadt...following @ipa-tif's spreadsheet Verbauliste.xlsx

The ultimate goal could be to even generate launch files based on a module configuration using a tool based on @ipa-mdl's roslaunch-dump

[fmessmer]

@ipa-mdl asked for input about roslaunch-dump (with the final goal being added to roslaunch)
ros/ros_comm#1022

[floweisshardt]

reminder for @ipa-fxm: extract sick scanners from base urdf + then remove nav configs for mlr bases as they do not have laser scanners anymore

[fmessmer]

extract sick scanners from base urdf + then remove nav configs for mlr bases as they do not have laser scanners anymore

done in #654

[fmessmer]

atm none of the solutions proposed by @ipa-fxm and @ipa-nhg fully meet our requirements
also current efforts require a final decision regarding future maintaining responsibilities and supported hardware

thus, I'm moving it to Backlog

[floweisshardt]

we won't work on this for now

[mathias-luedtke]

we won't work on this for now

No need to close it.

[fmessmer]

it's been in the sprint backlog anyway...