Enable samplerobot sample in kinetic

[Naoki-Hiraoka]

In kinetic, we have to set use_velocity_feedback to false (because #245).

After setting use_velocity_feedback to false, we have other problems in simulating samplebot.

roslaunch hrpsys_gazebo_general gazebo_samplerobot_no_controllers.launch
rtmlaunch hrpsys_gazebo_general samplerobot_hrpsys_bringup.launch

1. The robot vibrates.
2. Too many error messages arise. [hrpEC][1576049965.454544] Timeover: processing time = 2.38[ms] (but control rate is 200 Hz!)

To solve 1, I set p-gain to smaller value. 3bcc726
To solve 2, I had to modify /opt/ros/kinetic/share/hrpsys_ros_bridge/models/SampleRobot.RobotHardware.conf (not included in this PR)
from
exec_cxt.periodic.rate: 500
to
exec_cxt.periodic.rate: 200

(control rate is defined in 4 files

SampleRobot.RobotHardware.conf
(for execution and control of RobotHardware.rtc)

and

rtmros_gazebo/hrpsys_gazebo_general/launch/robot_hrpsys_bringup.launch

Line 23 in 7389a49

<arg name="HRPSYS_RATE" default="200" />

(for execution of other rtc)(default value)

and

rtmros_gazebo/hrpsys_gazebo_general/config/SampleRobot.conf

Line 2 in 7389a49

dt: 0.005

(for control of other rtc)

and

rtmros_gazebo/hrpsys_gazebo_general/src/IOBPlugin.cpp

Line 21 in 7389a49

iob_period(0.005),

(for IOBPlugin)(default value)
)

With these modification, the robot still cannot walk, so more debugging is needed.

[Naoki-Hiraoka]

Another bug was found.
fkanehiro/hrpsys-base#1288

[Naoki-Hiraoka]

SampleRobot walked in kinetic.

roslaunch hrpsys_gazebo_general gazebo_samplerobot_no_controllers.launch

rtmlaunch hrpsys_gazebo_general samplerobot_hrpsys_bringup.launch

roseus
load "package://hrpsys_ros_bridge/euslisp/samplerobot-interface.l
samplerobot-init
setq *robot* (instance SampleRobot-robot :init)
send *ri* :angle-vector (send *robot* :angle-vector) 5000
send *ri* :start-auto-balancer
send *ri* :start-st
send *ri* :go-pos 1 0 0 

The upper body vibrates while walking.
It seems that this is because the inertia of upper body of SampleRobot is very small so the simulation easily gets unstable.

[Naoki-Hiraoka]

This PR is included in #250