Welcome to the Multi-Agent Tracking Environment’s documentation!
Welcome to the documentation of MATE
, the Multi-Agent Tracking Environment.
The source code of the MultiAgentTracking
environment is hosted on GitHub.
You can find it at mate.
For detailed description, please checkout our paper (PDF, bibtex).
This is an asymmetric two-team zero-sum stochastic game with partial observations, and each team has multiple agents (multiplayer). Intra-team communications are allowed, but inter-team communications are prohibited. It is cooperative among teammates, but it is competitive among teams (opponents).
Installation
git config --global core.symlinks true # required on Windows
pip3 install git+https://github.com/XuehaiPan/mate.git#egg=mate
Note
Python 3.7+ is required, and Python versions lower than 3.7 is not supported.
It is highly recommended to create a new isolated virtual environment for MATE
using conda:
git clone https://github.com/XuehaiPan/mate.git && cd mate
conda env create --no-default-packages --file conda-recipes/basic.yaml # or full-cpu.yaml to install RLlib
conda activate mate
- Getting Started
- The Environment Details
- Built-in Wrappers
- Repeated Reward and Individual Done
- Enhanced Observation
- Shared Field of View
- More Training Information
- Relative Coordinates
- Rescaled Observation
- Discrete Action Spaces
- Single-Team Multi-Agent Setting
- Single-Team Single-Agent Setting
- Auxiliary Camera Rewards
- Auxiliary Target Rewards
- Message Filter
- Random Message Dropout
- Restricted Communication Range
- No Communication
- Extra Communication Delays
- Render Communication
- Modules
- mate package
- Subpackages
- Submodules
- mate.constants module
TERRAIN_SIZE
TERRAIN_WIDTH
TERRAIN_SPACE
WAREHOUSES
NUM_WAREHOUSES
WAREHOUSE_RADIUS
MAX_CAMERA_VIEWING_ANGLE
TARGET_RADIUS
PRESERVED_SPACE
PRESERVED_DIM
OBSERVATION_OFFSET
CAMERA_STATE_DIM_PUBLIC
CAMERA_STATE_SPACE_PUBLIC
CAMERA_STATE_DIM_PRIVATE
CAMERA_STATE_SPACE_PRIVATE
TARGET_STATE_DIM_PUBLIC
TARGET_STATE_SPACE_PUBLIC
TARGET_STATE_DIM_PRIVATE
TARGET_STATE_SPACE_PRIVATE
OBSTACLE_STATE_DIM
OBSTACLE_STATE_SPACE
CAMERA_ACTION_DIM
CAMERA_DEFAULT_ACTION
TARGET_ACTION_DIM
TARGET_DEFAULT_ACTION
camera_observation_space_of()
target_observation_space_of()
observation_space_of()
camera_observation_indices_of()
target_observation_indices_of()
observation_indices_of()
camera_observation_slices_of()
target_observation_slices_of()
observation_slices_of()
camera_coordinate_mask_of()
target_coordinate_mask_of()
coordinate_mask_of()
- mate.environment module
- mate.utils module
- Module contents
make()
camera_observation_space_of()
target_observation_space_of()
observation_space_of()
camera_observation_indices_of()
target_observation_indices_of()
observation_indices_of()
camera_observation_slices_of()
target_observation_slices_of()
observation_slices_of()
camera_coordinate_mask_of()
target_coordinate_mask_of()
coordinate_mask_of()
read_config()
EnvMeta
MultiAgentTracking
EnhancedObservation
SharedFieldOfView
RescaledObservation
RelativeCoordinates
MoreTrainingInformation
DiscreteCamera
DiscreteTarget
AuxiliaryCameraRewards
AuxiliaryTargetRewards
group_reset()
group_step()
group_observe()
group_communicate()
group_act()
MultiCamera
SingleCamera
MultiTarget
SingleTarget
MessageFilter
RestrictedCommunicationRange
RandomMessageDropout
NoCommunication
ExtraCommunicationDelays
RenderCommunication
RepeatedRewardIndividualDone
WrapperMeta
WrapperSpec
CameraAgentBase
TargetAgentBase
RandomCameraAgent
RandomTargetAgent
NaiveCameraAgent
NaiveTargetAgent
GreedyCameraAgent
GreedyTargetAgent
HeuristicCameraAgent
HeuristicTargetAgent
MixtureCameraAgent
MixtureTargetAgent
convert_coordinates()
normalize_observation()
rescale_observation()
split_observation()
CameraStatePublic
CameraStatePrivate
TargetStatePublic
TargetStatePrivate
ObstacleState
seed_everything()
sin_deg()
cos_deg()
tan_deg()
arcsin_deg()
arccos_deg()
arctan2_deg()
cartesian2polar()
polar2cartesian()
normalize_angle()
Vector2D
Team
Message
- mate package
Citation
If you find MATE useful, please consider citing:
@inproceedings{pan2022mate,
title = {{MATE}: Benchmarking Multi-Agent Reinforcement Learning in Distributed Target Coverage Control},
author = {Xuehai Pan and Mickel Liu and Fangwei Zhong and Yaodong Yang and Song-Chun Zhu and Yizhou Wang},
booktitle = {Thirty-sixth Conference on Neural Information Processing Systems Datasets and Benchmarks Track},
year = {2022},
url = {https://openreview.net/forum?id=SyoUVEyzJbE}
}