126 lines
5.1 KiB
Python
126 lines
5.1 KiB
Python
import random
|
|
from typing import Tuple, List, Union, Iterable
|
|
|
|
import numpy as np
|
|
from pathlib import Path
|
|
from environments import helpers as h
|
|
|
|
|
|
class BaseFactory:
|
|
|
|
@property
|
|
def movement_actions(self):
|
|
return (int(self.allow_vertical_movement) + int(self.allow_horizontal_movement)) * 4
|
|
|
|
def __init__(self, level='simple', n_agents=1, max_steps=1e3):
|
|
self.n_agents = n_agents
|
|
self.max_steps = max_steps
|
|
self.allow_vertical_movement = True
|
|
self.allow_horizontal_movement = True
|
|
self.level = h.one_hot_level(
|
|
h.parse_level(Path(__file__).parent / h.LEVELS_DIR / f'{level}.txt')
|
|
)
|
|
self.slice_strings = {0: 'level', **{i: f'agent#{i}' for i in range(1, self.n_agents+1)}}
|
|
self.reset()
|
|
|
|
def reset(self):
|
|
self.done = False
|
|
self.steps = 0
|
|
# Agent placement ...
|
|
agents = np.zeros((self.n_agents, *self.level.shape), dtype=np.int8)
|
|
floor_tiles = np.argwhere(self.level == h.IS_FREE_CELL)
|
|
# ... on random positions
|
|
np.random.shuffle(floor_tiles)
|
|
for i, (x, y) in enumerate(floor_tiles[:self.n_agents]):
|
|
agents[i, x, y] = h.IS_OCCUPIED_CELL
|
|
# state.shape = level, agent 1,..., agent n,
|
|
self.state = np.concatenate((np.expand_dims(self.level, axis=0), agents), axis=0)
|
|
# Returns State, Reward, Done, Info
|
|
return self.state, 0, self.done, {}
|
|
|
|
def additional_actions(self, agent_i, action) -> ((int, int), bool):
|
|
raise NotImplementedError
|
|
|
|
def step(self, actions):
|
|
actions = [actions] if isinstance(actions, int) else actions
|
|
assert isinstance(actions, list), f'"actions has to be in [{int, list}]'
|
|
self.steps += 1
|
|
|
|
# FixMe: Why do we need this?
|
|
r = 0
|
|
|
|
# Move this in a seperate function?
|
|
actions = list(enumerate(actions))
|
|
random.shuffle(actions)
|
|
for agent_i, action in actions:
|
|
if self._is_moving_action(action):
|
|
pos, valid = self.move_or_colide(agent_i, action)
|
|
else:
|
|
pos, valid = self.additional_actions(agent_i, action)
|
|
actions[agent_i] = (agent_i, action, pos, valid)
|
|
|
|
collision_vecs = self.check_all_collisions(actions, self.state.shape[0])
|
|
|
|
reward, info = self.calculate_reward(collision_vecs, [a[1] for a in actions], r)
|
|
r += reward
|
|
if self.steps >= self.max_steps:
|
|
self.done = True
|
|
return self.state, r, self.done, info
|
|
|
|
def _is_moving_action(self, action):
|
|
if action < self.movement_actions:
|
|
return True
|
|
else:
|
|
return False
|
|
|
|
def check_all_collisions(self, agent_action_pos_valid_tuples: (int, int, (int, int), bool), collisions: int) -> np.ndarray:
|
|
collision_vecs = np.zeros((len(agent_action_pos_valid_tuples), collisions)) # n_agents x n_slices
|
|
for agent_i, action, pos, valid in agent_action_pos_valid_tuples:
|
|
if self._is_moving_action(action):
|
|
collision_vecs[agent_i] = self.check_collisions(agent_i, pos, valid)
|
|
return collision_vecs
|
|
|
|
def check_collisions(self, agent_i: int, pos: (int, int), valid: bool) -> np.ndarray:
|
|
pos_x, pos_y = pos
|
|
# FixMe: We need to find a way to spare out some dimensions, eg. an info dimension etc... a[?,]
|
|
collisions_vec = self.state[:, pos_x, pos_y].copy() # "vertical fiber" at position of agent i
|
|
collisions_vec[h.AGENT_START_IDX + agent_i] = h.IS_FREE_CELL # no self-collisions
|
|
if valid:
|
|
# ToDo: Place a function hook here
|
|
pass
|
|
else:
|
|
collisions_vec[h.LEVEL_IDX] = h.IS_OCCUPIED_CELL
|
|
return collisions_vec
|
|
|
|
def do_move(self, agent_i: int, old_pos: (int, int), new_pos: (int, int)) -> None:
|
|
(x, y), (x_new, y_new) = old_pos, new_pos
|
|
self.state[agent_i + h.AGENT_START_IDX, x, y] = h.IS_FREE_CELL
|
|
self.state[agent_i + h.AGENT_START_IDX, x_new, y_new] = h.IS_OCCUPIED_CELL
|
|
|
|
def move_or_colide(self, agent_i: int, action: int) -> ((int, int), bool):
|
|
old_pos, new_pos, valid = h.check_agent_move(state=self.state,
|
|
dim=agent_i + h.AGENT_START_IDX,
|
|
action=action)
|
|
if valid:
|
|
# Does not collide width level boundrys
|
|
self.do_move(agent_i, old_pos, new_pos)
|
|
return new_pos, valid
|
|
else:
|
|
# Agent seems to be trying to collide in this step
|
|
return old_pos, valid
|
|
|
|
def agent_i_position(self, agent_i):
|
|
return tuple(np.argwhere(self.state[h.AGENT_START_IDX+agent_i] == h.IS_OCCUPIED_CELL).flatten())
|
|
|
|
@property
|
|
def free_cells(self) -> np.ndarray:
|
|
free_cells = self.state.sum(0)
|
|
free_cells = np.argwhere(free_cells == h.IS_FREE_CELL)
|
|
np.random.shuffle(free_cells)
|
|
return free_cells
|
|
|
|
def calculate_reward(self, collisions_vec: np.ndarray, actions: Iterable[int], r: int) -> (int, dict):
|
|
# Returns: Reward, Info
|
|
# Set to "raise NotImplementedError"
|
|
return 0, {}
|