MuhammadOsamaNusrat/rl-gridworld
1
1import numpy as np
2
3class GridWorld:
4 def __init__(self, max_steps=50):
5 self.size = 4
6 self.start = (0, 0)
7 self.goal = (3, 3)
8 self.holes = [(1, 1), (1, 3), (3, 0)]
9 self.actions = ['up', 'down', 'left', 'right']
10 self.max_steps = max_steps
11 self.reset()
12
13 def reset(self):
14 self.state = self.start
15 self.steps = 0
16 return self.state
17
18 def step(self, action):
19 x, y = self.state
20 self.steps += 1
21
22 if action == 'up': x = max(0, x - 1)
23 elif action == 'down': x = min(self.size - 1, x + 1)
24 elif action == 'left': y = max(0, y - 1)
25 elif action == 'right': y = min(self.size - 1, y + 1)
26
27 self.state = (x, y)
28
29 if self.state in self.holes:
30 return self.state, -50, True
31 elif self.state == self.goal:
32 return self.state, 50, True
33 elif self.steps >= self.max_steps:
34 return self.state, -10, True # episode timeout
35 else:
36 return self.state, -1, False
37
38 