Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
..
Failed to load latest commit information.
README.md Updated documentation to python example readme + scenarios readme (#189) Apr 16, 2017
basic.cfg Corrected configs and python examples. Oct 15, 2016
basic.wad Scenarios moved from examples dir to separate dir in root (restored) Apr 29, 2016
bots.cfg Moved configs to scenarios dir. Oct 15, 2016
cig.cfg Updated CIG Python examples Jul 2, 2018
cig.wad Corrected cig2 map texture bug. Jun 25, 2016
cig_with_unknown.wad Added unknown maps from cig2016 competition. Feb 2, 2017
deadly_corridor.cfg Corrected configs and python examples. Oct 15, 2016
deadly_corridor.wad Removed print from deadly corridor. May 10, 2016
deathmatch.cfg Corrected configs and python examples. Oct 15, 2016
deathmatch.wad ChainsawMarine increases killcount id defend_the_center and deathmatc… Jul 23, 2017
defend_the_center.cfg Corrected configs and python examples. Oct 15, 2016
defend_the_center.wad ChainsawMarine increases killcount id defend_the_center and deathmatc… Jul 23, 2017
defend_the_line.cfg Corrected configs and python examples. Oct 15, 2016
defend_the_line.wad Scenarios moved from examples dir to separate dir in root (restored) Apr 29, 2016
health_gathering.cfg Corrected configs and python examples. Oct 15, 2016
health_gathering.wad Scenarios moved from examples dir to separate dir in root (restored) Apr 29, 2016
health_gathering_supreme.cfg Added config for hgs. Jan 25, 2018
health_gathering_supreme.wad Changing superhealth for the las time. Apr 28, 2016
learning.cfg Corrected configs and python examples. Oct 15, 2016
multi.cfg Corrected configs and python examples. Oct 15, 2016
multi_deathmatch.wad Scenarios moved from examples dir to separate dir in root (restored) Apr 29, 2016
multi_duel.cfg Merged with current master. Nov 25, 2016
multi_duel.wad Imprved multiple_instances example May 28, 2016
my_way_home.cfg Corrected configs and python examples. Oct 15, 2016
my_way_home.wad Scenarios moved from examples dir to separate dir in root (restored) Apr 29, 2016
oblige.cfg Fix oblige.py example, update basic examples for all languages Apr 8, 2018
predict_position.cfg Corrected configs and python examples. Oct 15, 2016
predict_position.wad Changed predict position scenario. May 9, 2016
rocket_basic.cfg Corrected configs and python examples. Oct 15, 2016
rocket_basic.wad Changed missing doom2.wad textures in simpler basic and rocket basic … Jun 19, 2017
simpler_basic.cfg Corrected configs and python examples. Oct 15, 2016
simpler_basic.wad Changed missing doom2.wad textures in simpler basic and rocket basic … Jun 19, 2017
take_cover.cfg Corrected configs and python examples. Oct 15, 2016
take_cover.wad Scenarios moved from examples dir to separate dir in root (restored) Apr 29, 2016

README.md

Scenarios Decription:

Scenarios contained in iwad files do not support action constraints, death penalty and living rewards. Every mention of any settings that are not included in iwads will be specified with "(config)".

Note: Vizdoom does not support setting certain rewards (such as killing oponents) in .cfg files. These must be set in the .wad files instead

BASIC

The purpose of the scenario is just to check if using this framework to train some AI i 3D environment is feasible.

Map is a rectangle with gray walls, ceiling and floor. Player is spawned along the longer wall, in the center. A red, circular monster is spawned randomly somewhere along the opposite wall. Player can only (config) go left/right and shoot. 1 hit is enough to kill the monster. Episode finishes when monster is killed or on timeout.

REWARDS:

+101 for killing the monster -5 for missing Episode ends after killing the monster or on timeout.

Further configuration:

  • living reward = -1,
  • 3 available buttons: move left, move right, shoot (attack)
  • timeout = 300

DEADLY CORRIDOR

The purpose of this scenario is to teach the agent to navigate towards his fundamental goal (the vest) and make sure he survives at the same time.

Map is a corridor with shooting monsters on both sides (6 monsters in total). A green vest is placed at the oposite end of the corridor. Reward is proportional (negative or positive) to change of the distance between the player and the vest. If player ignores monsters on the sides and runs straight for the vest he will be killed somewhere along the way. To ensure this behavior doom_skill = 5 (config) is needed.

REWARDS:

+dX for getting closer to the vest. -dX for getting further from the vest.

Further configuration:

  • 5 available buttons: turn left, turn right, move left, move right, shoot (attack)
  • timeout = 4200
  • death penalty = 100
  • doom_skill = 5

DEFEND THE CENTER

The purpose of this scenario is to teach the agent that killing the monsters is GOOD and when monsters kill you is BAD. In addition, wasting amunition is not very good either. Agent is rewarded only for killing monsters so he has to figure out the rest for himself.

Map is a large circle. Player is spawned in the exact center. 5 melee-only, monsters are spawned along the wall. Monsters are killed after a single shot. After dying each monster is respawned after some time. Episode ends when the player dies (it's inevitable becuse of limitted ammo).

REWARDS: +1 for killing a monster

Further configuration:

  • 3 available buttons: turn left, turn right, shoot (attack)
  • death penalty = 1

DEFEND THE LINE

The purpose of this scenario is to teach the agent that killing the monsters is GOOD and when monsters kill you is BAD. In addition, wasting amunition is not very good either. Agent is rewarded only for killing monsters so he has to figure out the rest for himself.

Map is a rectangle. Player is spawned along the longer wall, in the center. 3 melee-only and 3 shooting monsters are spawned along the oposite wall. Monsters are killed after a single shot, at first. After dying each monster is respawned after some time and can endure more damage. Episode ends when the player dies (it's inevitable becuse of limitted ammo).

REWARDS: +1 for killing a monster

Further configuration:

  • 3 available buttons: turn left, turn right, shoot (attack)
  • death penalty = 1

HEALTH GATHERING

The purpose of this scenario is to teach the agent how to survive without knowing what makes him survive. Agent know only that life is precious and death is bad so he must learn what prolongs his existence and that his health is connected with it.

Map is a rectangle with green, acidic floor which hurts the player periodically. Initially there are some medkits spread uniformly over the map. A new medkit falls from the skies every now and then. Medkits heal some portions of player's health - to survive agent needs to pick them up. Episode finishes after player's death or on timeout.

Further configuration:

  • living_reward = 1
  • 3 available buttons: turn left, turn right, move forward
  • 1 available game variable: HEALTH
  • death penalty = 100

MY WAY HOME

The purpose of this scenario is to teach the agent how to navigate in a labirynth-like surroundings and reach his ultimate goal (and learn what it actually is).

Map is a series of rooms with interconnection and 1 corridor with a dead end. Each room has a different color. There is a green vest in one of the rooms (the same room every time). Player is spawned in randomly choosen room facing a random direction. Episode ends when vest is reached or on timeout/

REWARDS: +1 for reaching the vest

Further configuration:

  • 3 available buttons: turn left, turn right, move forward
  • living reward = -0.0001
  • timeout = 2100

PREDICT POSITION

The purpose of the scenario is teach agent to synchronize missle weapon shot (involving a signifficant delay between shooting and hitting) with target movements. Agent should be able to shoot so that missle and monster meet each other.

The map is a rectangle room. Player is spawned along the longer wall, in the center. A monster is spawned randomly somewhere along the opposite wall and walks between left and right corners along the wall. Player is equipped with a rocket launcher and a single rocket. Episode ends when missle hits a wall/the monster or on timeout.

REWARDS: +1 for killing the monster

Further configuration:

  • living reward = -0.0001,
  • 3 available buttons: move left, move right, shoot (attack)
  • timeout = 300

TAKE COVER

The purpose of this scenario is to teach agent to link incomming missles with his estimated lifespan. Agent should learn that being hit means health decrease and this in turn will lead to death which is undesirable. In effect agent should avoid missles.

Map is a rectangle. Player is spawned along the longer wall, in the center. A couple of shooting monsters are spawned randomly somewhere along the opposite wall and try to kill the player with fireballs. The player can only (config) move left/right. More monsters appear with time. Episode ends when player dies.

REWARDS: +1 for each tic of life

Further configuration:

  • living reward = 1.0,
  • 2 available buttons: move left, move right