# Copyright (c) 2020, Fabio Muratore, Honda Research Institute Europe GmbH, and
# Technical University of Darmstadt.
# All rights reserved.
#
# Redistribution and use in source and binary forms, with or without
# modification, are permitted provided that the following conditions are met:
# 1. Redistributions of source code must retain the above copyright
# notice, this list of conditions and the following disclaimer.
# 2. Redistributions in binary form must reproduce the above copyright
# notice, this list of conditions and the following disclaimer in the
# documentation and/or other materials provided with the distribution.
# 3. Neither the name of Fabio Muratore, Honda Research Institute Europe GmbH,
# or Technical University of Darmstadt, nor the names of its contributors may
# be used to endorse or promote products derived from this software without
# specific prior written permission.
#
# THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND
# ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
# WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
# DISCLAIMED. IN NO EVENT SHALL FABIO MURATORE, HONDA RESEARCH INSTITUTE EUROPE GMBH,
# OR TECHNICAL UNIVERSITY OF DARMSTADT BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
# SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO,
# PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS;
# OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER
# IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
# ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
# POSSIBILITY OF SUCH DAMAGE.
import random
from copy import deepcopy
import numpy as np
from pyrado.environments.pysim.base import SimEnv
from pyrado.spaces.base import Space
from pyrado.utils.data_types import RenderMode
[docs]class MockEnv(SimEnv):
"""
A mock version of SimEnv, used in env wrapper tests.
State, observation, action, and init spaces as well as the task are passed to the constructor.
The step() method saves the last action into the last_act attribute. The action value is converted to a list to
ease assertions. step() and reset() return next_obs as observation. If it is None, a random vector is returned.
"""
def __init__(self, state_space=None, obs_space=None, act_space=None, init_space=None, task=None):
# Set spaces
self._state_space = state_space
self._obs_space = obs_space
self._act_space = act_space
self._init_space = init_space
self._task = task
# Set empty domain param
self._domain_param = {}
self._dt = 1.0
# Init check attributes
self.next_obs = None
self.next_reward = None
self.next_step_done = False
self.last_act = None
@property
def obs_space(self):
if self._obs_space is None:
raise NotImplementedError
return self._obs_space
@property
def state_space(self):
if self._state_space is None:
# Use observation space if no state space was specified (for backwards compatibility)
return self._obs_space
return self._state_space
@property
def act_space(self):
if self._act_space is None:
raise NotImplementedError
return self._act_space
@property
def init_space(self):
if self._init_space is None:
raise NotImplementedError
return self._init_space
@init_space.setter
def init_space(self, space: Space):
# No checks for MockEnv
self._init_space = space
@property
def task(self):
raise self._task
def _create_task(self, task_args: dict = None):
pass # unused
@property
def domain_param(self) -> dict:
return deepcopy(self._domain_param)
@domain_param.setter
def domain_param(self, domain_param):
self._domain_param.clear()
self._domain_param.update(domain_param)
[docs] def get_nominal_domain_param(self):
return {}
def _get_obs(self):
# Return None if no obs space
if self._obs_space is None:
return None
# Return random if no next_obs set
if self.next_obs is None:
return self._obs_space.sample_uniform()
return np.array(self.next_obs)
[docs] def reset(self, init_state=None, domain_param=None):
# Set domain params
if domain_param is not None:
self.domain_param = domain_param
# Return next observation
return self._get_obs()
[docs] def step(self, act: np.ndarray) -> tuple:
# Store as last action as list, to simplify asserts
if self._act_space is not None:
self.last_act = list(act)
# Return next observation
obs = self._get_obs()
# And next reward
rew = self.next_reward
if rew is None:
rew = random.random()
return obs, rew, self.next_step_done, dict()
[docs] def render(self, mode=RenderMode(), render_step=1):
# No visual
pass