change to gym-control dir
pip install -e .
change to gymtest dir
python main.py
the map:
[0 0 0 1 100 1 0 0 0 0]
The goal is to find one tile where the agent gets 100 chocolate He can speed up to get there faster
since he either speeds up or down, he cant stay one one tile
He can get more rewards if he stays in that area
Observation is a pair of (position, speed), where the position can be 0 -> length of rewardtable, speed can range from -10 to 10
There are two actions he can make speed or slow by 1 unit