Hello, I pushed some python environments for Multi Agent Reinforcement Learning.

Python 100.00%

multi-agent-reinforcement-learning-environment's Introduction

Multi-Agent-Learning-Environments

Hello, I pushed some python environments for Multi Agent Reinforcement Learning. Some are single agent version that can be used for algorithm testing. I provide documents for each environment, you can check the corresponding pdf files in each directory. These are just toy problems, though some of them are still hard to solve. Some environments are like:

Multi Agent Soccer Game

Multi Agent Rescue

Multi Agent Cleaner

Multi Agent Move Box

Multi Agent Catching Pig

Multi Drones Monitoring

Multi Agent Maze Running

Multi Agent Find Treasure

Firefighters

Go Together

Warehouse

Opposite

Dependency

OpenCV, swig

Multi-Agent Environment Standard

Assumption:

Each agent works synchronously.

Member Functions

reset()

reward_list, done = step(action_list)

obs_list = get_obs()

reward_list records the single step reward for each agent, it should be a list like [reward1, reward2,......]. The length should be the same as the number of agents. Each element in the list should be a integer.

done True/False, mark when an episode finishes.

action_list records the single step action instruction for each agent, it should be a list like [action1, action2,...]. The length should be the same as the number of agents. Each element in the list should be a non-negative integer.

obs_list records the single step observation for each agent, it should be a list like [obs1, obs2,...]. The length should be the same as the number of agents. Each element in the list can be any form of data, but should be in same dimension, usually a list of variables or an image.

Typical Monte Carlo Procedures

reset environment by calling reset() get initial observation get_obs() for i in range(max_MC_iter): get action_list from controller apply action by step() record returned reward list record new observation by get_obs()

Citation

Cite the environment of the following paper as:

@inproceedings{jiang2021multi,
 title={Multi-agent reinforcement learning with directed exploration and selective memory reuse},
 author={Jiang, Shuo and Amato, Christopher},
 booktitle={Proceedings of the 36th Annual ACM Symposium on Applied Computing},
 pages={777--784},
 year={2021}
}

multi-agent-reinforcement-learning-environment's People

Contributors

Stargazers

Watchers

Forkers

wanghuimu yuejin-ee wu850952384 w470062742 hassamsheikh siddharthghiya awesome-archive frankfan007 landoufulxf zhangnyg wwxfromtju george88888 ssinghzar abluceli senshuran robotsl zhuyifengzju hyyh28 sangwonseo94 verystrongjoe bhargavaram1997 chenkehan21 hi-researcher xuannadi huhaoran jiang1206 tzuren bututoubaobei kalman61 dlml ecustboy indra-ipd lyu-xg nivekhah deepmind007 zcchenvy theodoreoor mehdizouitine xuanphu108 curieuxjy tianqi-777 qw7798965 zhangran2006 tchen00 mieximiemie wwwliuliuliu laozhanger tang-core faithmai yskim525 zhaowei1997 freesia-l benellis3 bekercai bengreenberg5 mgilgamesh pgkang butterfly2sea aicools zhong-shan wangxinwi rafaelmp2 chunjiangmonkey rainwangphy piccupl alirezashamsoshoara viviankeith jiangbowangfr justinhedge idomatheveryday jy1001 seulbinhwang zxnuli magicianlial hsinglukliu neilwth harrison12315 luofuliang rodrigoclira exilesaber valauri skydvn clarenceluo78 harry-zhou amirezzati xudongzhang-de miingyang eboyu gp-chen lonelyanger lsb567 mrx1a wxyrc5 zhiyizeng ralami1859 nov-17th wqq0921 yuling91 hiuhiuwong matejdurech

multi-agent-reinforcement-learning-environment's Issues

博主，你有女朋友吗

Issue while running env_Rescue

I am facing the following error on Mac can you help me out:
Traceback (most recent call last):
File "test.py", line 2, in
from env_rescue import EnvRescue
File "/Users/ashwinnair/PycharmProjects/Multi-Agent-Reinforcement-Learning-Environment/env_Rescue/Python3/env_rescue.py", line 1, in
from CEnvRescue import CEnvRescue
File "/Users/ashwinnair/PycharmProjects/Multi-Agent-Reinforcement-Learning-Environment/env_Rescue/Python3/CEnvRescue.py", line 20, in
_CEnvRescue = swig_import_helper()
File "/Users/ashwinnair/PycharmProjects/Multi-Agent-Reinforcement-Learning-Environment/env_Rescue/Python3/CEnvRescue.py", line 18, in swig_import_helper
return importlib.import_module('_CEnvRescue')
File "/Users/ashwinnair/opt/anaconda3/lib/python3.7/importlib/init.py", line 127, in import_module
return _bootstrap._gcd_import(name[level:], package, level)
ImportError: dlopen(/Users/ashwinnair/PycharmProjects/Multi-Agent-Reinforcement-Learning-Environment/env_Rescue/Python3/_CEnvRescue.so, 2): no suitable image found. Did find:
/Users/ashwinnair/PycharmProjects/Multi-Agent-Reinforcement-Learning-Environment/env_Rescue/Python3/_CEnvRescue.so: unknown file type, first eight bytes: 0x7F 0x45 0x4C 0x46 0x02 0x01 0x01 0x00
/Users/ashwinnair/PycharmProjects/Multi-Agent-Reinforcement-Learning-Environment/env_Rescue/Python3/_CEnvRescue.so: unknown file type, first eight bytes: 0x7F 0x45 0x4C 0x46 0x02 0x01 0x01 0x00

Under what circumstances will the env_FireFighter be done?

Thank you for your great work, I'm a new learner of the RL field, and I'm learning how to build my model.

for other environments, I can see how the game finish, such as agents put the box in the right location. but in the FireFighter environment, when will an episode be done?

I have tried to let fire level == [0,0,0,0] (or [2,2,2,2]) be the goal, at this time I will end the episode and give a positive reward. I use a DQN to learn the strategy, but it seems that whatever actions I choose, the fire level will increase especially the first and last house. I wonder how I can set the stopping criterion in this environment, do you have any idea? thank you!

Will you provide continuous actions for all envs?

Hi, this repo is very interesting. Will you provide continuous actions for all envs?

License ?

Under what license are these reinforcement learning environments released under ? Thank you.

Have you publish any papers based on the environments?

Hi! This repo is very interesting Have you publish any papers that use the environments?

Regarding the env_Soccer

Hi. Thanks for putting up all these envs. It is really helpful for a person like me who is new to both RL and Python,

It will be very helpful if you can let me know what are the local rewards to each agent for every action. The reason why I am asking is that the overall objective is to score the goals (and defend them by the opponent team), but there must be some incentive for each agent to perform the given actions, or is the reward +10 and -10 sufficient?

I am eagerly waiting for your response. Thanks.

bigpig4396 / multi-agent-reinforcement-learning-environment Goto Github PK