不久前,师弟问了我一个问题:“师兄,我要控制一个机器人在仿真环境下运动,需要学什么,或者从哪里入手呢?“
这个问题不是那么好回答,urdf——tf——Gazebo——ros_control——MoveIt,这是我在十多天后才能给出的一个答案。那么之前我是怎么做仿真的呢?
1. 借助Arbotix舵机接口来仿真
之前是在学习《ros by example volume 2 》的时候,接触到一个rbx2_bringup 的package,仿照着里面的https://github.com/pirobot/rbx2/blob/indigo-devel/rbx2_bringup/launch/pi_robot_with_gripper.launch
文件写了一个关于youbot的关节的,另外加上一些static_tf转换。
内容如下
<launch> <!-- Make sure we are not using simulated time --> <param name="/use_sim_time" value="false" /> <!-- Launch the arbotix driver in fake mode by default --> <arg name="sim" default="true" /> <!-- If using a real controller, look on /dev/ttyUSB0 by default --> <arg name="port" default="/dev/ttyUSB0" /> <!-- Load the URDF/Xacro model of our robot --> <param name="robot_description" command="$(find xacro)/xacro.py '$(find youbot_description)/robots/youbot.urdf.xacro'" /> <param name="use_gui" value="true"/> <node name="rviz" pkg="rviz" type="rviz" /> <!-- <include file="$(find urdf_tutorial)/launch/display.launch" />--> <!-- Bring up the arbotix driver with a configuration file appropriate to the robot --> <node name="arbotix" pkg="arbotix_python" type="arbotix_driver" clear_params="true" output="screen"> <rosparam file="$(find rbx2_bringup)/config/fake_youbot_arbotix.yaml" command="load" /> <param name="sim" value="$(arg sim)" /> <param name="port" value="$(arg port)" /> </node> <!-- Run a separate controller for the one sided gripper --> <node name="right_gripper_controller" pkg="arbotix_controllers" type="gripper_controller" output="screen"> <rosparam> model: dualservo min_opening: 0.0 max_opening: 0.009 invert_left: false invert_right: false center_left: 0.0 center_right: 0.0 pad_width: 0.015 finger_length: 0.04 joint_left: gripper_finger_joint_l joint_right: gripper_finger_joint_r </rosparam> </node> <!-- Publish the robot state --> <node name="robot_state_publisher" pkg="robot_state_publisher" type="state_publisher"> <param name="publish_frequency" type="double" value="20.0" /> </node> <!-- <node name="joint_state_publisher" pkg="joint_state_publisher" type="joint_state_publisher" />--> <!-- Start all servos in a relaxed state --> <node pkg="rbx2_dynamixels" type="arbotix_relax_all_servos.py" name="relax_all_servos" unless="$(arg sim)" /> <!-- Load diagnostics --> <node pkg="diagnostic_aggregator" type="aggregator_node" name="diagnostic_aggregator" clear_params="true" unless="$(arg sim)"> <rosparam command="load" file="$(find rbx2_dynamixels)/config/dynamixel_diagnostics.yaml" /> </node> <node pkg="rqt_robot_monitor" type="rqt_robot_monitor" name="rqt_robot_monitor" unless="$(arg sim)" /><!-- Unless value evaluates to false then include tag and its contents,inverse with if --> <!-- http://wiki.ros.org/tf#static_transform_publisher--> <!--yaw is rotation about Z, pitch is rotation about Y, and roll is rotation about X frame_id child_frame_id--> <!-- Run a static transform between /base_link and /base_footprint needed for SLAM --> <node pkg="tf" type="static_transform_publisher" name="wheel_link_fl2base_link" args="0.228 0.158 -0.034 0 0 0 /base_link /wheel_link_fl 100" /> <node pkg="tf" type="static_transform_publisher" name="wheel_link_fr2base_link" args="0.228 -0.158 -0.034 0 0 0 /base_link /wheel_link_fr 100" /> <node pkg="tf" type="static_transform_publisher" name="wheel_link_bl2base_link" args="-0.228 0.158 -0.034 0 0 0 /base_link /wheel_link_bl 100" /> <node pkg="tf" type="static_transform_publisher" name="wheel_link_br2base_link" args="-0.228 -0.158 -0.034 0 0 0 /base_link /wheel_link_br 100" /> <node pkg="tf" type="static_transform_publisher" name="caster_link_fl2base_link" args="0.228 0.158 -0.034 0 0 0 /base_link /caster_link_fl 100" /> <node pkg="tf" type="static_transform_publisher" name="caster_link_fr2base_link" args="0.228 -0.158 -0.034 0 0 0 /base_link /caster_link_fr 100" /> <node pkg="tf" type="static_transform_publisher" name="caster_link_bl2base_link" args="-0.228 0.158 -0.034 0 0 0 /base_link /caster_link_bl 100" /> <node pkg="tf" type="static_transform_publisher" name="caster_link_bf2base_link" args="-0.228 -0.158 -0.034 0 0 0 /base_link /caster_link_br 100" /> </launch>
它加载的yaml文件内容如下
port: /dev/ttyUSB0 baud: 115200 rate: 20 sync_write: True sync_read: True read_rate: 20 write_rate: 20 joints: { # wheel_joint_fl: {id: 1, neutral: 512, max_speed: 684, min_angle: 0, max_angle: 360, invert: False}, # wheel_joint_fr: {id: 2, neutral: 512, max_speed: 684, min_angle: 0, max_angle: 360, invert: False}, # wheel_joint_bl: {id: 3, neutral: 512, max_speed: 684, min_angle: 0, max_angle: 360, invert: False}, # wheel_joint_br: {id: 4, neutral: 512, max_speed: 684, min_angle: 0, max_angle: 360, invert: False}, wheel_joint_fl: {id: 1, neutral: 512, max_speed: 684, invert: False}, wheel_joint_fr: {id: 2, neutral: 512, max_speed: 684, invert: False}, wheel_joint_bl: {id: 3, neutral: 512, max_speed: 684, invert: False}, wheel_joint_br: {id: 4, neutral: 512, max_speed: 684, invert: False}, arm_joint_1: {id: 5, neutral: 512, max_speed: 684, min_angle: -169, max_angle: 169, invert: False}, arm_joint_2: {id: 6, neutral: 512, max_speed: 684, min_angle: -65, max_angle: 90, invert: False}, arm_joint_3: {id: 7, neutral: 512, max_speed: 684, min_angle: -151, max_angle: 146, invert: False}, arm_joint_4: {id: 8, neutral: 512, max_speed: 684, min_angle: -102.5, max_angle: 102.5, invert: False}, arm_joint_5: {id: 9, neutral: 512, max_speed: 684, min_angle: -167.5, max_angle: 167.5, invert: False}, gripper_finger_joint_l: {id: 10,neutral: 512, max_speed: 684, invert: False}, gripper_finger_joint_r: {id: 11,neutral: 512, max_speed: 684, invert: False} } #http://wiki.ros.org/arbotix_python controllers: { # Pololu motors: 1856 cpr = 0.3888105m travel = 4773 ticks per meter (empirical: 4684) WheelRadius_[meter] = 0.0475 EncoderTicksPerRound = 4000 base_controller: {type: diff_controller, base_frame_id: base_link, base_width: 0.38, ticks_meter: 4684, Kp: 50, Kd: 0, Ki: 20, Ko: 50, accel_limit: 1.0 }, arm_controller: {type: follow_controller, joints: [arm_joint_1, arm_joint_2, arm_joint_3, arm_joint_4, arm_joint_5], action_name: arm_1/arm_controller/follow_joint_trajectory} # rbx2/rbx2_dynamixels/config/arbotix/pi_robot_with_gripper.yaml # arm_controller: {type: follow_controller, joints: [right_arm_shoulder_pan_joint, right_arm_shoulder_lift_joint, right_arm_shoulder_roll_joint, right_arm_elbow_flex_joint, right_arm_forearm_flex_joint], action_name: arm_1/arm_controller/follow_joint_trajectory} }
启动后,运行起来的节点图
如果用这个仿真,很大程度上是有问题的,虽然也能够让模型动起来,主要是因为Arbotix是针对舵机的,而实际中很多的机器人不只是舵机,也有步进电机等等。我们的KUKA youBot机械臂是步进电机。以前的仿真中主要用的是follow_joint_trajectory action topics。而实际的youbot_driver也会启动这个topic,所以格式方面,仿真中能够用的,实际也可以直接用上,最大的问题是,仿真的参数与实际动作的相似度有有多少,这里是有个很大的问号的。
2. 借助Moveit
关于Moveit之前研究过一段时间,也总结成了三篇博客。
在利用命令roslaunch moveit_setup_assistant setup_assistant.launch 配置好一些config文件后。要控制实际的机器人,我们还得自己设置一个~/catkin_ws/src/moveit_youbot/config/controllers.yaml的配置文件,我的配置如下
controller_list: - name: arm_1/arm_controller action_ns: follow_joint_trajectory type: FollowJointTrajectory default: true joints: - arm_joint_1 - arm_joint_2 - arm_joint_3 - arm_joint_4 - arm_joint_5
从这里也可以看出,它也是用的Action接口来达到控制的目的的。搜索过一些资料后发现好像还真没其他的配置的方法,基本都是用action来做的。
- https://github.com/ros-planning/moveit_plugins/tree/c52c84dbb8764e8b995f10ab99654c4f65f8335b/moveit_ros_control_interface
- http://answers.ros.org/question/71824/gazebo-and-moveit-moveitsimplecontrollermanager-and-sending-joint-trajectory-messages/
- http://answers.ros.org/question/84806/using-moveit-to-actually-control-a-robot/
- https://groups.google.com/forum/#!topic/moveit-users/zNRG50tHuuY
其实根据youbot实际的Driver
我还尝试过这样的配置,为的是只使用/arm_1/arm_controller/position_command这个topic,而非action。也有网上的资料说那个type是ros_controller的type,而它又是与Gazebo相关的。在这里我设置的是topic 的type
controller_list: - name: arm_1/arm_controller/position_command type: brics_actuator/JointPositions default: true joints: - arm_joint_1 - arm_joint_2 - arm_joint_3 - arm_joint_4 - arm_joint_5
~/catkin_ws/src/moveit_youbot/launchyoubot_moveit_controller_manager.launch.xml 则如下:
<launch> <arg name="moveit_controller_manager" default="moveit_simple_controller_manager/MoveItSimpleControllerManager"/> <!-- <arg name="moveit_controller_manager" default="moveit_ros_control_interface/MoveItControllerManager"/>--> <param name="moveit_controller_manager" value="$(arg moveit_controller_manager)"/> <rosparam file="$(find moveit_youbot)/config/controllers.yaml"/> </launch>
貌似也只能使用MoveItSimpleControllerManager了。根据这里的说明点击打开链接。它提供action的接口,绕了个路还是回到了action。也就是这种方式貌似是行不通。
3. Gazebo与ros_control
很久之前学习Gazebo的时候,那个官方tutorial的rrbot,2自由度连杆机器人,结合rqt的一些工具联合调pid,现在依然印象深刻,视频地址点击打开链接。不过当时没太深入,现在随着积累的东西多了,慢慢又会回头看看以前的东西,总有新的收获。而这次则主要是关于ros_control。这里我会先跑一遍官方的例子,然后结合youBot的相关文件来对比说明。
整理了一下,基本上按照这几个地方就好,按照顺序跑一遍:
- http://gazebosim.org/tutorials?tut=ros_roslaunch
- http://gazebosim.org/tutorials/?tut=ros_urdf
- http://gazebosim.org/tutorials?tut=ros_gzplugins
- http://gazebosim.org/tutorials/?tut=ros_control
其中涉及urdf的相关知识,具体参考查看roswiki, http://wiki.ros.org/urdf/XML。
官方package的目录结构如下
这里,我主要是想说说plugin和ros_control,介绍可以看看http://wiki.ros.org/ros_control,开头的那个PPT很不错,可以有个大概了解。
ros_control主要是提供各种controller和Hardware硬件抽象层,而实体机器人或者仿真部件就是其控制的资源,它用来连接它们
下面就结合具体的官方代码来分析下,是如何让仿真中的模型动起来的。首先~/catkin_ws/src/gazebo_ros_demos/rrbot_description/urdf/rrbot.xacro文件定义了两个运动关节的transmission,它的type是:transmission_interface/SimpleTransmission,它的hardwareInterface是:EffortJointInterface
<transmission name="tran1"> <type>transmission_interface/SimpleTransmission</type> <joint name="joint1"> <hardwareInterface>EffortJointInterface</hardwareInterface> </joint> <actuator name="motor1"> <hardwareInterface>EffortJointInterface</hardwareInterface> <mechanicalReduction>1</mechanicalReduction> </actuator> </transmission>
接着~/catkin_ws/src/gazebo_ros_demos/rrbot_description/urdf/rrbot.gazebo里有一些传感器的plugin:libgazebo_ros_control.so,libgazebo_ros_gpu_laser.so,libgazebo_ros_camera.so。hokuyo激光雷达的如下,也就是说即使没有传感器,也可以仿真得到传感器的数据。
<gazebo reference="hokuyo_link"> <sensor type="gpu_ray" name="head_hokuyo_sensor"> <pose>0 0 0 0 0 0</pose> <visualize>false</visualize> <update_rate>40</update_rate> <ray> <scan> <horizontal> <samples>720</samples> <resolution>1</resolution> <min_angle>-1.570796</min_angle> <max_angle>1.570796</max_angle> </horizontal> </scan> <range> <min>0.10</min> <max>30.0</max> <resolution>0.01</resolution> </range> <noise> <type>gaussian</type> <!-- Noise parameters based on published spec for Hokuyo laser achieving "+-30mm" accuracy at range < 10m. A mean of 0.0m and stddev of 0.01m will put 99.7% of samples within 0.03m of the true reading. --> <mean>0.0</mean> <stddev>0.01</stddev> </noise> </ray> <plugin name="gazebo_ros_head_hokuyo_controller" filename="libgazebo_ros_gpu_laser.so"> <topicName>/rrbot/laser/scan</topicName> <frameName>hokuyo_link</frameName> </plugin> </sensor> </gazebo>
关于ros_control的plugin也在这个文件中,如下
<gazebo> <plugin name="gazebo_ros_control" filename="libgazebo_ros_control.so"> <robotNamespace>/rrbot</robotNamespace> <robotSimType>gazebo_ros_control/DefaultRobotHWSim</robotSimType> </plugin> </gazebo>
关于它们的说明,如下图
Controllers:
- effort_controllers
- joint_effort_controller
- joint_position_controller
- joint_velocity_controller
- joint_state_controller
- joint_state_controller
- position_controllers
- joint_position_controller
- velocity_controllers
- joint_velocity_controllers
Hardware Interface:
- joint Command Interfaces
- Effort Joint Interface
- Velocity Joint Interface
- Position Joint Interface
- Joint State Interfaces
- Actuator State Interfaces
- Actuator Command Interfaces
- Effort Actuator Interface
- Velocity Actuator Interface
- Position Actuator Interface
- Joint Pose Trajectory
- Differential Drive
- Skid Steering Drive(Pioneer 3AT for instance)
- Plannar Move plugin
- Bumper
- GPU Laser(e.g. Hokuyo)
- Block Laser(e.g. Velodyne)
- Force-torque sensor Interface
- IMU sensor Interface
ros_control又是和urdf中的transmission相关的,这里有部分解释http://wiki.ros.org/ros_control 具体语法还可以参考http://wiki.ros.org/urdf/XML/Transmission
依照官网的解释ros_control会去读取~/catkin_ws/src/gazebo_ros_demos/rrbot_description/urdf/rrbot.xacro中的transmisson条目,每个运动关节都需要一个transmission。
那么这些运动关节是以什么方式运动,或者pid参数是如何设定呢?因此就需要一个配置文件来配置这些。这里名叫~/catkin_ws/src/gazebo_ros_demos/rrbot_control/config/rrbot_control.yaml文件。Hardwareinterface的作用就在这个配置文件中体现了,注意Joint1和Joint2的type
rrbot: # Publish all joint states ----------------------------------- joint_state_controller: type: joint_state_controller/JointStateController publish_rate: 50 # Position Controllers --------------------------------------- joint1_position_controller: type: effort_controllers/JointPositionController joint: joint1 pid: {p: 100.0, i: 0.01, d: 10.0} joint2_position_controller: type: effort_controllers/JointPositionController joint: joint2 pid: {p: 100.0, i: 0.01, d: 10.0}
具体的关于启动的文件,都写在launch文件中了,具体来说就是rrbot_world.launch和rrbot_control.launch,内容分别如下
<launch> <!-- these are the arguments you can pass this launch file, for example paused:=true --> <arg name="paused" default="false"/> <arg name="use_sim_time" default="true"/> <arg name="gui" default="true"/> <arg name="headless" default="false"/> <arg name="debug" default="false"/> <!-- We resume the logic in empty_world.launch, changing only the name of the world to be launched --> <include file="$(find gazebo_ros)/launch/empty_world.launch"> <arg name="world_name" value="$(find rrbot_gazebo)/worlds/rrbot.world"/> <arg name="debug" value="$(arg debug)" /> <arg name="gui" value="$(arg gui)" /> <arg name="paused" value="$(arg paused)"/> <arg name="use_sim_time" value="$(arg use_sim_time)"/> <arg name="headless" value="$(arg headless)"/> </include> <!-- Load the URDF into the ROS Parameter Server --> <param name="robot_description" command="$(find xacro)/xacro.py '$(find rrbot_description)/urdf/rrbot.xacro'" /> <!-- Run a python script to the send a service call to gazebo_ros to spawn a URDF robot --> <node name="urdf_spawner" pkg="gazebo_ros" type="spawn_model" respawn="false" output="screen" args="-urdf -model rrbot -param robot_description"/> </launch>
以下的这个lanuch文件是用来启动controller_manager和发布joint states
<launch> <!-- Load joint controller configurations from YAML file to parameter server --> <rosparam file="$(find rrbot_control)/config/rrbot_control.yaml" command="load"/> <!-- load the controllers --> <node name="controller_spawner" pkg="controller_manager" type="spawner" respawn="false" output="screen" ns="/rrbot" args="joint_state_controller joint1_position_controller joint2_position_controller"/> <!-- convert joint states to TF transforms for rviz, etc --> <node name="robot_state_publisher" pkg="robot_state_publisher" type="robot_state_publisher" respawn="false" output="screen"> <remap from="/joint_states" to="/rrbot/joint_states" /> </node> </launch>
上面的这几个文件都是比较通用的,可以当作参考的模板。
以下的这个链接是总的一个关于ROS和Gazebo的系列教程。
http://gazebosim.org/tutorials?cat=connect_ros