mirror of
https://github.com/Farama-Foundation/Gymnasium.git
synced 2025-07-31 22:04:31 +00:00
e913bc81b8
* feat: add `isort` to `pre-commit` * ci: skip `__init__.py` file for `isort` * ci: make `isort` mandatory in lint pipeline * docs: add a section on Git hooks * ci: check isort diff * fix: isort from master branch * docs: add pre-commit badge * ci: update black + bandit versions * feat: add PR template * refactor: PR template * ci: remove bandit * docs: add Black badge * ci: try to remove all `|| true` statements * ci: remove lint_python job - Remove `lint_python` CI job - Move `pyupgrade` job to `pre-commit` workflow * fix: avoid messing with typing * docs: add a note on running `pre-cpmmit` manually * ci: apply `pre-commit` to the whole codebase
58 lines
1.8 KiB
Python
58 lines
1.8 KiB
Python
import numpy as np
|
|
|
|
from gym import utils
|
|
from gym.envs.mujoco import mujoco_env
|
|
|
|
|
|
class AntEnv(mujoco_env.MujocoEnv, utils.EzPickle):
|
|
def __init__(self):
|
|
mujoco_env.MujocoEnv.__init__(self, "ant.xml", 5)
|
|
utils.EzPickle.__init__(self)
|
|
|
|
def step(self, a):
|
|
xposbefore = self.get_body_com("torso")[0]
|
|
self.do_simulation(a, self.frame_skip)
|
|
xposafter = self.get_body_com("torso")[0]
|
|
forward_reward = (xposafter - xposbefore) / self.dt
|
|
ctrl_cost = 0.5 * np.square(a).sum()
|
|
contact_cost = (
|
|
0.5 * 1e-3 * np.sum(np.square(np.clip(self.sim.data.cfrc_ext, -1, 1)))
|
|
)
|
|
survive_reward = 1.0
|
|
reward = forward_reward - ctrl_cost - contact_cost + survive_reward
|
|
state = self.state_vector()
|
|
notdone = np.isfinite(state).all() and state[2] >= 0.2 and state[2] <= 1.0
|
|
done = not notdone
|
|
ob = self._get_obs()
|
|
return (
|
|
ob,
|
|
reward,
|
|
done,
|
|
dict(
|
|
reward_forward=forward_reward,
|
|
reward_ctrl=-ctrl_cost,
|
|
reward_contact=-contact_cost,
|
|
reward_survive=survive_reward,
|
|
),
|
|
)
|
|
|
|
def _get_obs(self):
|
|
return np.concatenate(
|
|
[
|
|
self.sim.data.qpos.flat[2:],
|
|
self.sim.data.qvel.flat,
|
|
np.clip(self.sim.data.cfrc_ext, -1, 1).flat,
|
|
]
|
|
)
|
|
|
|
def reset_model(self):
|
|
qpos = self.init_qpos + self.np_random.uniform(
|
|
size=self.model.nq, low=-0.1, high=0.1
|
|
)
|
|
qvel = self.init_qvel + self.np_random.standard_normal(self.model.nv) * 0.1
|
|
self.set_state(qpos, qvel)
|
|
return self._get_obs()
|
|
|
|
def viewer_setup(self):
|
|
self.viewer.cam.distance = self.model.stat.extent * 0.5
|