`BHV_Agent`

Note

This page will not explain anything about the basics of MOOS-IvP. If you are new and looking for resources on this checkout their homepage. Consider completing the labs associated with MIT 2.680 if you are really serious in your goal to learn MOOS-IvP.

Note

BHV_Agent has only been tested in scenarios where:

It has the maximum pwt out of all behaviors in the mission it is used
Has perpetual = true set in it's configuration block

Okay fine, a little MOOS-IvP primer

Behaviors, are the "IvP" part of MOOS-IvP which take in "MOOS" variables and produce "IvP Functions" which dictate how the robots running MOOS-IvP should behavior. For a good primer see these 2.680 Lecture Slides (2021).

Purpose of BHV_Agent

In the case of moos-ivp-agent, BHV_Agent is forwarding all "MOOS state" to the python ModelBridgeServer (wrapped by MissionManager). The python ML models then make decisions on what action a robot running MOOS-IvP and BHV_Agent should take by providing a course and speed and optionally mutate the "MOOS State" of the robot by making posts.

Configuration

Note

See the general Helm Autonomy reference for information about configuring behaviors in general.

BHV_Agent exports three types of state from MOOS-IvP to python

(Automatic): The position of the robot
- As defined by the NAV_X and NAV_Y moos vars.
(Upon subscription): The position of other robots.
- As supplied by the NODE_REPORT_<VNAME> and pNodeParse (aquaticus)
(Upon subscription): Any specified MOOS var

Subscribing to vehicle positions

The sub_vehicle configuration key is used to subscribe to vehicles based on their name. The following BHV_Agent config block shows a subscription to two vehicles with names evan and felix.

Behavior = BHV_Agent
{
  name = agent
  pwt  = 100
  perpetual = true

  condition = MODE == ACTIVE

  sub_vehicle = evan
  sub_vehicle = felix
}

Subscribing to MOOS vars

The sub_var configuration key is used to subscribe to MOOS vars based on their state. The following BHV_Agent config block shows a subscription to two vars with names TAGGED and HAS_FLAG.

Note

As some MOOS-IvP apps have startup delays or do not post vars initially, you cannot assume that vars subscribed to will be immediately available on the python side. You are responsible for checking before you access.

Behavior = BHV_Agent
{
  name = agent
  pwt  = 100
  perpetual = true

  condition = MODE == ACTIVE

  sub_var = TAGGED
  sub_var = HAS_FLAG
}

What does state look like in Python

Once the state is outputted from MissionManager, it will be in a python dictionary with the following form. All values will be translated into python's string, float, or boolean types (With the exception of the node report dictionaries).

>>> msg = mgr.get_message()
>>> print(msg.state)
{
  'NAV_X': 30.5,
  'NAV_Y': -75.2,
  'TAGGED': False,
  'HAS_FLAG': True,
  'NODE_REPORTS': {
    'evan': {
      'NAV_X': -60.3
      'NAV_Y': 35.2
    },
    'felix': {
      'NAV_X': -12.3
      'NAV_Y': -54.0
    }
  }
}

BHV_Agent and pEpisodeManager

BHV_Agent will automatically collect data from pEpisodeManager if it is present and forward that data to the python model. See the (MissionMessage)[api_pages/mission_message.md] for information on how to access this.