STRANDS Executive

Old documentation, being updated now (from 9.2.205)

Executive control code for STRANDS robots. The basic unit of behaviour is a task as defined by strands_executive_msgs/Task. To get the robot to execute a task an appropriate instance of the Task message must be sent to the task executor framework.

Dependencies

Notes, before you can run any of the scheduling stuff, you must run have the datacentre running, e.g.

roslaunch mongodb_store mongodb_store.launch

or with path specifing, where should db be stored:

roslaunch mongodb_store mongodb_store.launch db_path:=/...

and you need to be offering a 'topological_navigation', GotoNodeAction action. If you're not running the full topological navigation system, you can run

roslaunch topological_utils dummy_topological_navigation.launch

which will provide a fake map. TODO: more information about locations in map

Running scheduled patrols

To test the executive framework you can try running the robot around the topological map. To do this, first get your basic 2D navigation setup running (in simulation or reality). Next, build a topological map. Then (assuming you have no special transitions in your map), run the monitored navigation plus the topological navigation:

roslaunch monitored_navigation monitored_nav.launch
roslaunch topological_navigation topological_navigation.launch map:=<topological_map_name> node_by_node:=false

Then you can start the executive framework:

roslaunch task_executor task-scheduler.launch

With this up and running you can start the robot running continuous patrols using:

rorun task_executor continuous_patrolling.py

Creating a Task

Most task instances will contain both the name of a topological map node where the task should be executed, plus the name of a SimpleActionServer to be called at the node and its associated arguments. Tasks must contain one of these things, but not necessarily both (just a node name will just take the robot there, just an action will execute the action without moving the robot).

To create a task, first create an instance of the Task message type. Examples are given in Python, as the helper functions currently only exist for Python, but C++ is also possible (and C++ helpers will be added soon).

from strands_executive_msgs.msg import Task
task = Task()

Then you can set the node id for where the task will be executed (or you can do this inline):

task.start_node_id = 'WayPoint1'

To add the name of the action server, do:

task.action = 'do_dishes'

Where 'do_dishes' is replaced by the first argument you would give to the actionlib.SimpleActionClient constructor.

You should also set the expected duration of the task. This will be used by the scheduler and also by the execution framework to determine whether it things your task may be misbehaving. The duration is a rospy.Duration instance and is defined in seconds.

dishes_duration = 60 * 60
task.max_duration = rospy.Duration(dishes_duration)

If your actionlib goal type needs arguments, you must then add them to the task in the order they are used in your goal type constructor. You can either add plain string arguments (which are stored in the task itself) or ROS message instances (which are stored in the mongodb_store). For example, for the following action which is available under task_executor/action/TestExecution.action

# something to print
string some_goal_string
# something to test typing
geometry_msgs/Pose test_pose
# something for numbers
int32 an_int
float32 a_float
---
---
# feedback message
float32 percent_complete

You need to supply a string argument followed by a pose, then an int then a float. To add the string, do the following

from strands_executive_msgs import task_utils
task_utils.add_string_argument(task, 'my string argumment goes here')

For the pose, this must be added to the mongodb_store message store and then the ObjectID of the pose is used to communicate its location. This is done as follows

p = Pose()
object_id = msg_store.insert(p)
task_utils.add_object_id_argument(task, object_id, Pose)

Ints and floats can be added as follows

task_utils.add_int_argument(task, 24)
task_utils.add_float_argument(task, 63.678)

Registering a Task

A single task can be registered with the task executor and started:

add_task_srv_name = '/task_executor/add_task'
set_exe_stat_srv_name = '/task_executor/set_execution_status'
rospy.wait_for_service(add_task_srv_name)
rospy.wait_for_service(set_exe_stat_srv_name)
add_task_srv = rospy.ServiceProxy(add_task_srv_name, AddTask)
set_execution_status = rospy.ServiceProxy(set_exe_stat_srv_name, SetExecutionStatus)
    
try:
	# add task to the execution framework
    task_id = add_task_srv(task)
    # make sure execution is running -- this only needs to be done onece      
    set_execution_status(True)
except rospy.ServiceException, e: 
	print "Service call failed: %s"%e

Creating a Routine

The scenario use case for task execution is that the robot has a daily routine which is a list of tasks which it carries out every day. This can be created with the task_routine.DailyRoutine object which is configured with start and end times for the robot's daily activities:

	# some useful times
    localtz = tzlocal()
    # the time the robot will be active
    start = time(8,30, tzinfo=localtz)
    end = time(17,00, tzinfo=localtz)
    midday = time(12,00, tzinfo=localtz)

    morning = (start, midday)
    afternoon = (midday, end)

    routine = task_routine.DailyRoutine(start, end)

Tasks are then added using the repeat_every* methods. These take the given task and store it such that it can be correctly instantiated with start and end times every day:

	# do this task every day
    routine.repeat_every_day(task)
    # and every two hours during the day
    routine.repeat_every_hour(task, hours=2)
    # once in the morning
    routine.repeat_every(task, *morning)
    # and twice in the afternoon
    routine.repeat_every(task, *afternoon, times=2)

The DailyRoutine is a static thing (just a bunch of convenience functions really). The routine tasks are passed to the DailyRoutineRunner which manages the creation of specific task instances according to the schedule, and then sends them to the scheduler before they need to be added to teh schedule.