## repository: https://github.com/danielsnider/lost_comms_recovery
<<PackageHeader(lost_comms_recovery)>>

<<TOC>>

== Overview ==
If your robot loses connection to the base station it will navigate to a configurable home. The base station connection check uses ping to a configurable list of IPs.

== Quick Start ==
1. Install:

{{{
$ sudo apt-get install ros-kinetic-lost-comms-recovery
}}}

2. Launch:

{{{
$ roslaunch lost_comms_recovery lost_comms_recovery.launch ips_to_monitor:=192.168.1.2
}}}

== Details ==

This node is not real-time safe. The monitoring loop waits 3 seconds between checks. 

=== If move_base is running ===

If move_base is running, an autonomous recovery navigation will take place. The default position of the recovery goal is the origin (0,0) of the frame given it the goal_frame_id param and the orientation is all 0s by default. This default pose can be overridden if a messaged is published on the recovery_pose topic. If move_base is already navigating to a goal it will not be interrupted and recovery navigation will happen when move_base is idle.

=== If move_base is not running ===

If move_base is not running when communication failure occurs then motors and joysticks are set to zero by publishing a zero `geometry_msgs/Twist` message and a zero `sensor_msgs/Joy` message.

'''Important''': Instead of relying on this this feature, you're better off with motor control software that sets zero velocity after a certain amount of time not receiving any new commands.

== Nodes ==
{{{
#!clearsilver CS/NodeAPI
name = lost_comms_recovery

pub {
  0.name = cmd_vel
  0.type = geometry_msgs/Twist
  0.desc = Send a zero `Twist` message on this topic when communication to the base station is lost.
  1.name = joy
  1.type = sensor_msgs/Joy
  1.desc = Send a zero `Joy` message on this topic when communication to the base station is lost.
  }
sub {
  0.name = recovery_pose
  0.type = geometry_msgs/PoseWithCovarianceStamped
  0.desc = A pose that the robot will recover to if communication to the base station is lost. This will override the default recovery goal which is to the origin (0,0) of the `goal_frame_id` param.
}
param {
  0.name = ~goal_frame_id
  0.type = string
  0.desc = The [[tf]] frame for `move_base` goals.
  0.default = map
  1.name = ~ping_fail_count
  1.type = int
  1.desc = Number of pings that must fail for communication to be considered lost. Each ping waits 1 second for a response.
  1.default = 2
  2.name = ~ips_to_monitor
  2.type = string
  2.desc = Comma separated list of IPs to monitor using ping.
  2.default = 192.168.1.2,192.168.1.3
}
act_called {
  0.name = move_base
  0.type = move_base_msgs/MoveBaseAction
  0.desc = The move_base action server performs the autonomous recovery navigation.
}
}}}

== Use Case ==
=== Use in the University Rover Competition (URC) ===
This functionality is helpful for the autonomous task of the University Rover Competition (URC) and we hope this helps you.

== Normal Output ==
{{{
$ roslaunch lost_comms_recovery lost_comms_recovery.launch ips_to_monitor:=192.168.190.136

[INFO] Monitoring base station on IP(s): 192.168.190.136.
[INFO] Connected to base station.
[INFO] Connected to base station.
...
[ERROR] No connection to base station.
[INFO] Executing move_base goal to position (x,y) 0.0, 0.0.
[INFO] Initial goal status: PENDING
[INFO] This goal has been accepted by the simple action server
[INFO] Final goal status: SUCCEEDED
[INFO] Goal reached.
}}}