pepper_dcm_robot

DCM stack integrating tools to control Pepper robot

Installation

• install dependencies

sudo apt-get install ros-indigo-pepper-robot ros-indigo-pepper-meshes ros-indigo-pepper-control ros-indigo-naoqi-dcm-driver

• then, install pepper_dcm_bringup or compile it from source

sudo apt-get install ros-indigo-pepper-dcm-bringup

• optionally, install pepper_moveit_config

sudo apt-get install ros-indigo-pepper-moveit-config

How to use it

Trajectory control

To command your robot remotely with ROS control:

• start the DCM bringup proving your robot's IP (be aware that the package will stop Autonomous Life on your robot)

roslaunch pepper_dcm_bringup pepper_bringup.launch robot_ip:=<ROBOT_IP>

• start Naoqi Driver (to get the odom frame) proving your robot's IP

roslaunch naoqi_driver naoqi_driver.launch nao_ip:=<ROBOT_IP>

Control via MoveIt

• start Moveit! to control the robot via GUI

roslaunch pepper_moveit_config moveit_planner.launch

• check "Allow approximate IK solutions"
• control a planning_group via an interactive marker

Control via Actionlib

• you can send a trajectory to the desired controller (actionlib)

rosrun actionlib axclient.py <name of the goal topic of the action server>

example:

rosrun actionlib axclient.py /pepper_dcm/LeftArm_controller/follow_joint_trajectory/goal

To choose the controllers you want to load, modify pepper_control/launch/pepper_control_trajectory.launch. The list of implemented controllers, you can find in pepper_control/config/pepper_trajectory_control.yaml. You can start and stop the ros-controllers using the rqt plugin ControllerManager.

Position control

To command joints positions via ROS:

• start the DCM bringup proving your robot's IP (be aware that the package will stop Autonomous Life on your robot):

roslaunch pepper_dcm_bringup pepper_dcm_bringup_position.launch robot_ip:=<ROBOT_IP>

• send a position to the desired controller, for example

rostopic pub /pepper_dcm/HeadYaw_position_controller/command std_msgs/Float64 "data: 1"