habitat.tasks.rearrange.social_nav.social_nav_task.PddlSocialNavTask class

Social nav task based on PddlTask for training low-level policies under multi-agent setting

Methods

def _duplicate_sensor_suite(self, sensor_suite: core.simulator.SensorSuite) -> None
Modifies the sensor suite in place to duplicate articulated agent specific sensors between the two articulated agents.
def _generate_nav_start_goal(self, episode, force_idx = None) -> sub_tasks.nav_to_obj_task.NavToInfo
Returns the starting information for a navigate to object task.
def add_perf_timing(self, *args, **kwargs)
def get_action_name(self, action_index: typing.Union[int, numpy.integer])
def get_coll_forces(self, articulated_agent_id)
def get_cur_collision_info(self, agent_idx) -> utils.CollisionDetails
def get_n_targets(self) -> int
def overwrite_sim_config(self, config: typing.Any, episode: core.dataset.Episode) -> typing.Any
def reset(self, episode: core.dataset.Episode)
def seed(self, seed: int) -> None
def set_args(self, **kwargs)
def set_sim_reset(self, sim_reset)
def step(self, action: typing.Dict[str, typing.Any], episode: core.dataset.Episode)

Special methods

def __init__(self, *args, config, dataset = None, **kwargs)

Properties

abs_targ_idx get
action_space: gym.spaces.space.Space get
desired_resting get
initial_robot_trans get set
is_episode_active get
should_end: bool get set
targ_idx get