An environment for training agents that perform complete sweep exploration while learning to move at the same time
To run the environment just import the environment files:
from GoalEnvironment import GoalEnvironment
from GridEnvironment import GridEnvironment
env = GoalEnvironment()
env = GridEnvironment()
s = env.reset()
a = env.step(env.action_space.sample())
The GridEnvironment is the environment used for validation of the complete exploration
The GoalEnvironment may be used for training.
In order to configure obstacles just add json files to the obstacles folder.
For more information please refer to https://drive.google.com/file/d/1oXPZzeguVCWX0lIh9jegFPW9GlIUnVIz/view?usp=sharing