documentation config files

marl_factory_grid/configs/clean_and_bring.yaml

@@ -1,4 +1,21 @@
+General:
+  # RNG-seed to sample the same "random" numbers every time, to make the different runs comparable.
+  env_seed: 69
+  # Individual vs global rewards
+  individual_rewards: true
+  # The level.txt file to load from marl_factory_grid/levels
+  level_name: rooms
+  # Radius of Partially observable Markov decision process
+  pomdp_r: 3
+  # Print all messages and events
+  verbose: True
+  # Run tests
+  tests: false
+
+# In the "clean and bring" Scenario one agent aims to pick up all items and drop them at drop-off locations while all
+# other agents aim to clean dirt piles.
 Agents:
+  # The clean agents
   Wolfgang:
     Actions:
       - Move8
@@ -12,6 +29,7 @@ Agents:
       - DirtPiles
     Clones: 8
 
+  # The item agent
   Juergen:
     Actions:
       - Move8
@@ -42,33 +60,32 @@ Entities:
   Items:
     coords_or_quantity: 5
 
-
-General:
-  env_seed: 69
-  individual_rewards: true
-  level_name: rooms
-  pomdp_r: 3
-  verbose: True
-  tests: false
-
+# Rules section specifies the rules governing the dynamics of the environment.
 Rules:
   # Environment Dynamics
+  # When stepping over a dirt pile, entities carry a ratio of the dirt to their next position
   EntitiesSmearDirtOnMove:
     smear_ratio: 0.2
+  # Doors automatically close after a certain number of time steps
   DoorAutoClose:
     close_frequency: 7
 
   # Respawn Stuff
+  # Define how dirt should respawn after the initial spawn
   RespawnDirt:
     respawn_freq: 30
+  # Define how items should respawn after the initial spawn
   RespawnItems:
     respawn_freq: 50
 
   # Utilities
+  # Define what happens on entity collisions
   WatchCollisions:
     done_at_collisions: false
 
   # Done Conditions
+  # Define the conditions for the environment to stop. Either success or a fail conditions.
+  # The environment stops when all dirt is cleaned
   DoneOnAllDirtCleaned:
   DoneAtMaxStepsReached:
     max_steps: 500

marl_factory_grid/configs/default_config.yaml

@@ -1,5 +1,19 @@
 # Default Configuration File
 
+General:
+  # RNG-seed to sample the same "random" numbers every time, to make the different runs comparable.
+  env_seed: 69
+  # Individual vs global rewards
+  individual_rewards: true
+  # The level.txt file to load from marl_factory_grid/levels
+  level_name: large
+  # View Radius; 0 = full observatbility
+  pomdp_r: 3
+  # Print all messages and events
+  verbose: false
+  # Run tests
+  tests: false
+
 # Agents section defines the characteristics of different agents in the environment.
 
 # An Agent requires a list of actions and observations.
@@ -93,38 +107,42 @@ Entities:
   # Zones: Entities representing zones in the environment.
   Zones: { }
 
-# General section includes general settings for the environment.
-General:
-  env_seed: 69
-  individual_rewards: true
-  level_name: large
-  pomdp_r: 3
-  verbose: false
-  tests: false
 
 # Rules section specifies the rules governing the dynamics of the environment.
 Rules:
   # Environment Dynamics
+  # When stepping over a dirt pile, entities carry a ratio of the dirt to their next position
   EntitiesSmearDirtOnMove:
     smear_ratio: 0.2
+  # Doors automatically close after a certain number of time steps
   DoorAutoClose:
     close_frequency: 10
+  # Maintainers move at every time step
   MoveMaintainers:
 
   # Respawn Stuff
+  # Define how dirt should respawn after the initial spawn
   RespawnDirt:
     respawn_freq: 15
+  # Define how items should respawn after the initial spawn
   RespawnItems:
     respawn_freq: 15
 
   # Utilities
+  # Define what happens on entity collisions
   WatchCollisions:
     done_at_collisions: false
 
   # Done Conditions
+  # Define the conditions for the environment to stop. Either success or a fail conditions.
+  # The environment stops when an agent reaches a destination
   DoneAtDestinationReach:
+  # The environment stops when all dirt is cleaned
   DoneOnAllDirtCleaned:
+  # The environment stops when a battery is discharged
   DoneAtBatteryDischarge:
+  # The environment stops when a maintainer reports a collision
   DoneAtMaintainerCollision:
+  # The environment stops after max steps
   DoneAtMaxStepsReached:
     max_steps: 500

marl_factory_grid/configs/eight_puzzle.yaml

@@ -1,11 +1,17 @@
 General:
   env_seed: 69
+  # Individual vs global rewards
   individual_rewards: true
+  # The level.txt file to load from marl_factory_grid/levels
   level_name: eight_puzzle
+  # View Radius; 0 = full observatbility
   pomdp_r: 0
+  # Print all messages and events
   verbose: True
+  # Run tests
   tests: false
 
+# In the "eight puzzle" there are 8 agents standing on a 3x3 map, each with a specific destination to reach.
 Agents:
   Wolfgang:
     Actions:
@@ -27,7 +33,9 @@ Agents:
       - Dennis
       - Karl-Heinz
       - Kevin
+    # multiple agents can not stand on the same location
     is_blocking_pos: true
+
 Entities:
   Destinations:
     # Let them spawn on closed doors and agent positions
@@ -54,7 +62,8 @@ Rules:
     done_at_collisions: false
 
   # Done Conditions
+  # Done when all agents are standing on the correct destination at the same time
   DoneAtDestinationReach:
-    condition: simultanious
+    condition: simultaneous
   DoneAtMaxStepsReached:
     max_steps: 500

marl_factory_grid/configs/narrow_corridor.yaml

@@ -1,14 +1,16 @@
 General:
   # Your Seed
   env_seed: 69
-  # Individual or global rewards?
+  # Individual vs global rewards
   individual_rewards: true
-  # The level.txt file to load
+  # The level.txt file to load from marl_factory_grid/levels
   level_name: narrow_corridor
   # View Radius; 0 = full observatbility
   pomdp_r: 0
   # print all messages and events
   verbose: true
+  # Run tests
+  tests: false
 
 Agents:
   # Agents are identified by their name 

marl_factory_grid/configs/two_rooms_one_door.yaml

@@ -1,30 +1,18 @@
 General:
   env_seed: 69
+  # Individual vs global rewards
   individual_rewards: true
+  # The level.txt file to load from marl_factory_grid/levels
   level_name: two_rooms
+  # View Radius; 0 = full observatbility
   pomdp_r: 3
+  # Print all messages and events
   verbose: false
+  # Run tests
+  tests: false
 
-Entities:
-  Destinations: {}
-  Doors: {}
-  GlobalPositions: {}
-  Zones: {}
-
-Rules:
-  # Init:
-  AssignGlobalPositions: {}
-  ZoneInit: {}
-  AgentSingleZonePlacement: {}
-  IndividualDestinationZonePlacement: {}
-  # Env Rules
-  MaxStepsReached:
-    max_steps: 10
-  Collision:
-    done_at_collisions: false
-  DoorAutoClose:
-    close_frequency: 10
-
+# In "two rooms one door" scenario 2 agents spawn in 2 different rooms that are connected by a single door. Their aim
+# is to reach the destination in the room they didn't spawn in leading to a conflict at the door.
 Agents:
   Wolfgang:
     Actions:
@@ -49,3 +37,29 @@ Agents:
           - Walls
           - Destination
           - Doors
+
+Entities:
+  Destinations: { }
+  Doors: { }
+  GlobalPositions: { }
+  Zones: { }
+
+Rules:
+  # Environment Dynamics
+  DoorAutoClose:
+    close_frequency: 10
+
+  # Utilities
+  # Define what happens on entity collisions
+  WatchCollisions:
+    done_at_collisions: false
+
+  # Init
+  AssignGlobalPositions: { }
+  ZoneInit: { }
+  AgentSingleZonePlacement: { }
+  IndividualDestinationZonePlacement: { }
+
+  # Done Conditions
+  MaxStepsReached:
+    max_steps: 10

marl_factory_grid/modules/destinations/rules.py

@@ -13,7 +13,7 @@ from marl_factory_grid.utils.states import Gamestate
 
 ANY = 'any'
 ALL = 'all'
-SIMULTANEOUS = 'simultanious'
+SIMULTANEOUS = 'simultaneous'
 CONDITIONS = [ALL, ANY, SIMULTANEOUS]