Class Swim

dm_control/suite/fish.py:131–172

A Fish `Task` for swimming with smooth reward.

class Swim(base.Task):
  """A Fish `Task` for swimming with smooth reward."""

  def __init__(self, random=None):
    """Initializes an instance of `Swim`.

    Args:
      random: Optional, either a `numpy.random.RandomState` instance, an
        integer seed for creating a new `RandomState`, or None to select a seed
        automatically (default).
    """
    super().__init__(random=random)

  def initialize_episode(self, physics):
    """Sets the state of the environment at the start of each episode."""

    quat = self.random.randn(4)
    physics.named.data.qpos['root'][3:7] = quat / np.linalg.norm(quat)
    for joint in _JOINTS:
      physics.named.data.qpos[joint] = self.random.uniform(-.2, .2)
    # Randomize target position.
    physics.named.model.geom_pos['target', 'x'] = self.random.uniform(-.4, .4)
    physics.named.model.geom_pos['target', 'y'] = self.random.uniform(-.4, .4)
    physics.named.model.geom_pos['target', 'z'] = self.random.uniform(.1, .3)
    super().initialize_episode(physics)

  def get_observation(self, physics):
    """Returns an observation of joints, target direction and velocities."""
    obs = collections.OrderedDict()
    obs['joint_angles'] = physics.joint_angles()
    obs['upright'] = physics.upright()
    obs['target'] = physics.mouth_to_target()
    obs['velocity'] = physics.velocity()
    return obs

  def get_reward(self, physics):
    """Returns a smooth reward."""
    radii = physics.named.model.geom_size[['mouth', 'target'], 0].sum()
    in_target = rewards.tolerance(np.linalg.norm(physics.mouth_to_target()),
                                  bounds=(0, radii), margin=2*radii)
    is_upright = 0.5 * (physics.upright() + 1)
    return (7*in_target + is_upright) / 8