LIRRA  1.0.0
Loading...
Searching...
No Matches
rl_stl.envs.skydio_x2.hover Namespace Reference

Classes

class  HoverEnvConfig
 
class  ResetConfig
 
class  RewardConfig
 
class  RewardScales
 
class  STLRewardConfig
 
class  TerminationConfig
 
class  X2Hover
 

Functions

config_dict.ConfigDict to_config_dict (HoverEnvConfig cfg)
 
config_dict.ConfigDict default_config ()
 
 hover_box_formula (halfwidth, int window)
 
 periodic_exit_formula (halfwidth, int window)
 

Variables

str ENV_VERSION = "x2_hover_ctbr_v1"
 
dict STL_SPECS = {"hover_box": hover_box_formula, "periodic_exit": periodic_exit_formula}
 

Detailed Description

MuJoCo Playground (MJX/JAX) hover task for the Menagerie Skydio X2.

Action
    Normalized CTBR ``a = [T, p, q, r] in [-1, 1]^4`` (:mod:`.ctbr`). The action is
    held for ``n_substeps = ctrl_dt / sim_dt`` physics steps; the body-rate loop and
    the prioritized mixer run at every physics step on the measured body rates.

Observation (flat, float32; frames stated per block)
    - ``pos_err_world`` (3): ``p_target - p_com`` in world frame [m]; ``p_target`` is
      fixed (hover) or sampled per episode within ``target_range`` (waypoint variant)
    - ``vel_com_world`` (3): COM linear velocity, world frame [m/s]
    - ``rot_body_to_world`` (9): rotation matrix, row-major (encodes gravity direction and yaw)
    - ``ang_vel_body`` (3): body-frame angular velocity (gyro) [rad/s]
    - ``last_action`` (4)
    - ``context`` (3): ``[time_to_deadline / deadline, in_box_now, episode_time_fraction]``
      -- STL/task context slots (clock and instantaneous obligation predicate)
    - ``stl_window`` (``window * 3``, only if ``stl.enabled``): the last ``window``
      COM positions relative to the target, oldest first, so the windowed STL
      penalty below is a function of the observation (Markov).

Reward (task terms; each term is logged under ``reward/<name>``)
    ``pos`` = exp(-|e_p|^2 / sigma_pos^2), ``vel`` = exp(-|v|^2 / sigma_vel^2),
    ``upright`` = (1 + R_zz) / 2, ``ang_vel`` = |w|^2, ``action_rate`` = |a - a_prev|^2,
    ``alive`` = 1; ``reward = sum(scale_i * term_i)`` (penalty scales are negative).

Optional STL term (``stl.enabled``; disabled by default = task reward only)
    Specification (built with stlpy, selected by ``stl.spec``) over the signal
    ``y = p_com - p_target``:

    - ``hover_box`` (default): ``phi = G_[0, window-1] (y in box(halfwidth))`` --
      stay inside the box for the whole window;
    - ``periodic_exit``: ``phi = F_[0, window-1] (y not in box(halfwidth))`` -- leave
      the box at least once per window, i.e. never dwell inside it for ``window``
      consecutive samples. Scored at every step, this is the sliding-window form of
      ``G F_[0, window-1] (not in box)``.

    At control step ``k`` the realized window ``y_{k-window+1..k}`` is scored with the
    compiled JAX robustness and the step receives ``-stl_scale * max(0, delta - rho_k)``.
    ``rho_k`` is computed (and logged) whether or not the term is enabled, so a
    task-only policy can be checked against the same spec. Timing and credit
    assignment are explicit:

    - the term is **zero before** ``k_deadline + window - 1`` (``k_deadline =
      deadline_s / ctrl_dt``): the obligation holds from the deadline on, and only
      fully post-deadline windows are scored;
    - ``rho_k`` is a *past-time sliding-window* value: one sample that decides the
      window (an out-of-box sample for ``hover_box``, the only exit for
      ``periodic_exit``) enters the robustness of each of the ``window`` steps whose
      window contains it (credit is spread over the window, not assigned to a single
      transition);
    - no trajectory-level robustness is copied to all transitions.

Termination
    NaN/Inf in state or observation, tilt beyond ``max_tilt_deg``, COM below
    ``min_z``, or position error above ``max_pos_err``.

Floor contacts are disabled by default (``contacts=False``): the task terminates
above the floor, and disabling contacts removes collision cost from the batch.

Brax's ``BraxAutoResetWrapper`` with ``full_reset=False`` restores ``data``/``obs``
but not ``info``; the env therefore re-initializes its per-episode info whenever
``data.time == 0`` at the start of a step.

Function Documentation

◆ default_config()

config_dict.ConfigDict rl_stl.envs.skydio_x2.hover.default_config ( )

◆ hover_box_formula()

rl_stl.envs.skydio_x2.hover.hover_box_formula (   halfwidth,
int  window 
)
``G_[0, window-1] (|y_i| <= h_i for i = 0,1,2)`` over ``y = p_com - p_target``.

◆ periodic_exit_formula()

rl_stl.envs.skydio_x2.hover.periodic_exit_formula (   halfwidth,
int  window 
)
``F_[0, window-1] (|y_i| >= h_i for some i)`` over ``y = p_com - p_target``.

Robustness is ``max_{j in window} max_i (|y_{j,i}| - h_i)``: negative iff the COM was
strictly inside the box at every sample of the window.

◆ to_config_dict()

config_dict.ConfigDict rl_stl.envs.skydio_x2.hover.to_config_dict ( HoverEnvConfig  cfg)

Variable Documentation

◆ ENV_VERSION

str rl_stl.envs.skydio_x2.hover.ENV_VERSION = "x2_hover_ctbr_v1"

◆ STL_SPECS

dict rl_stl.envs.skydio_x2.hover.STL_SPECS = {"hover_box": hover_box_formula, "periodic_exit": periodic_exit_formula}