policy

SwarmNavigator

aymisxx · Stable-Baselines3

or hover any field below to flag it

Overview

Name
SwarmNavigator
Author
aymisxx
Framework
Stable-Baselines3
License
MIT
Skill type
navigation
Evidence level
untested
Task description
RL framework for multi-agent coverage & collision avoidance in 2D gridworlds. PPO/DQN train 1–4 agents; swarm covers 3x faster. Full visuals, GIFs, logs. Python // Gymnasium // SB3

Spaces

Action space
other · 0-dim · 0Hz
Observation space
  • type: other

Links

HuggingFace repo
null
Paper (arXiv)
null

Compatible robots

20

Compatible environments

0

No environments list SwarmNavigator yet.

Datasets that reference this policy

0

No datasets reference SwarmNavigator yet.