Ros导航虚拟墙设置

虚拟墙是在turtlebot2那里调包出来修改的，主要在RVIZ上面布置一些虚拟的墙体，并加进去虚拟的激光，好让虚拟的墙体也拥有膨胀区域，机器人到达虚拟墙体能够进行避障。

1、安装好yocs_virtual_sensor的包，这个包需要安装比较多的依赖文件

$ git clone https://github.com/yujinrobot/yujin_ocs.git
$ sudo apt-get install ros-kinetic-ecl-*
$ git clone https://github.com/yujinrobot/yocs_msgs.git
$ sudo apt-get install ros-kinetic-ar-track-alvar

2、编译yovs_virtual_sensor包，这个包需要自己更改ＣＭａｋｅLists.txt文件，不然启动launch文件时候会提示找不到节点，主要更改下面这几句：

## Specify additional locations of header files
include_directories(include ${catkin_INCLUDE_DIRS} ${Boost_INCLUDE_DIRS})
 
## Declare a cpp executable
add_executable(virtual_sensor_node src/virtual_sensor_node.cpp)
 
## Add cmake target dependencies of the executable/library
add_dependencies(virtual_sensor_node yocs_msgs_gencpp)
 
## Specify libraries to link a library or executable target against
target_link_libraries(virtual_sensor_node ${catkin_LIBRARIES})
 
#############
## Install ##
#############
 
install(TARGETS virtual_sensor_node
        DESTINATION ${CATKIN_PACKAGE_BIN_DESTINATION}
)

3、配置standalone.launch文件，需要添加yaml文件进去，还有虚拟墙的节点。

<launch>
  <arg name="virtual_wall" default="$(find yocs_virtual_sensor)/data/wall_list.yaml"/>
  <node name="virtual_sensor" pkg="yocs_virtual_sensor" type="virtual_sensor_node" >
    <param name="range_min"    value="0.0"/>
    <param name="range_max"    value="6.0"/>
    <param name="frequency"    value="10.0"/>
    <param name="hits_count"   value="3"/>
    <param name="global_frame" value="/map"/>
    <param name="sensor_frame" value="/base_link"/>
  </node>
　　<node pkg="tf" type="static_transform_publisher" name="base_to_virtual" args="0 0 0 0 0 0 /base_footprint /virtual_laser 50"/>
  <node name="wall_publisher" pkg="yocs_virtual_sensor" type="wall_publisher.py" required="true" >
    <param name="~filename" value="$(arg virtual_wall)"/>
  </node>
 
</launch>

４、添加好虚拟激光进去costmap.yaml文件，本来一个scan的，现在添加为两个scan。

obstacle_layer:
      enabled: true
      combination_method: 1
      track_unknown_space: true
      obstacle_range: 2.5
      raytrace_range: 3.0
      observation_sources: scan scan1
      scan: {
          sensor_frame: /laser_link,
          data_type: LaserScan,                                                                                                                                                                           
          topic: /scan,
          marking: true,
          clearing: true
      }
      scan1: {
          sensor_frame: /laser_link,
          data_type: LaserScan,
          topic: /virtual_sensor_scan,
          marking: true,
          clearing: false
      }

5、修改wall_publisher.py。

#修改
wall_pub   = rospy.Publisher('wall_pose_list', WallList,    latch = True)
#为
wall_pub   = rospy.Publisher('wall_pose_list', WallList,    latch = True,queue_size= 10)

6、可以启动导航包看效果了，后面墙的参数自己再好好消化。