unilab.envs.locomotion.common.dr_provider.LocomotionDRProvider

class unilab.envs.locomotion.common.dr_provider.LocomotionDRProvider[source]

Bases: DomainRandomizationProvider

Base DR provider for locomotion joystick environments.

Shared logic: - validate, build_interval_randomization_plan, _sample_commands - build_reset_plan (template with hooks) - build_reset_observation (template with hook)

Override these hooks in subclasses: - _get_qvel_limit — default 0.5 - _build_extra_info_updates — default empty dict - _compute_reset_obs — must be implemented per robot

Methods

__init__()

build_init_randomization_plan(env)

build_interval_randomization_plan(env, ...)

build_reset_observation(env, env_ids, ...)

build_reset_plan(env, env_ids)

validate(env, capabilities)

validate(env, capabilities)[source]
Parameters:
Return type:

None

build_interval_randomization_plan(env, step_counter)[source]
Parameters:

  • env (Any)

  • step_counter (int)

Return type:

IntervalRandomizationPlan | None

build_reset_plan(env, env_ids)[source]
Parameters:
Return type:

ResetPlan

build_reset_observation(env, env_ids, info_updates)[source]
Parameters:
Return type:

dict[str, ndarray]

build_init_randomization_plan(env)
Parameters:

env (Any)

Return type:

InitRandomizationPlan | None