unilab.tasks.locomotion.g1.manager_terms.G1WalkRewardPack

class unilab.tasks.locomotion.g1.manager_terms.G1WalkRewardPack[source]

Bases: ManagerTermBase

Evaluate the canonical walk reward family in one tensor carrier read.

This pack fuses the G1 owner’s state-only per-step rewards: linear/angular tracking, angular-xy and orientation penalties, action rate, weighted pose, foot-orientation penalty, and phase tracking. Equations, weights, and declaration order remain owned by RewardManager; this owner only removes duplicate carrier reads and intermediate per-term kernel launches.

Parameters:

Methods

__init__(cfg, env)

Attributes

last_step_timing_ms

name

num_envs

returns_transient_tensor

tensor_sensor_names

returns_transient_tensor = True
last_step_timing_ms: dict[str, float] = {}
__init__(cfg, env)[source]
Parameters:
property tensor_sensor_names: tuple[str, ...]
__call__(env)[source]

Returns the value of the term required by the manager.

Parameters:

env (_G1Env)

Return type:

Tensor

property name: str
property num_envs: int