unilab.terrains.heightfield_terrains.HfPyramidStairsTerrainCfg

class unilab.terrains.heightfield_terrains.HfPyramidStairsTerrainCfg[source]

Bases: SubTerrainCfg

A pyramid stairs terrain encoded as a heightfield.

Concentric square rings from the outside in form a staircase climbing toward a central platform. With holes=True the four diagonal corners of each ring are carved out to a deep pit; agents falling into the pit reach a terminating depth instead of an infinite void.

Parameters:

  • proportion (float)

  • size (tuple[float, float])

  • flat_patch_sampling (dict[str, FlatPatchSamplingCfg] | None)

  • step_height_range (tuple[float, float])

  • step_width (float)

  • platform_width (float)

  • border_width (float)

  • holes (bool)

  • pit_depth (float)

  • horizontal_scale (float)

  • vertical_scale (float)

  • base_thickness_ratio (float)

Methods

__init__([proportion, size, ...])

function(difficulty, rng)

Generate backend-agnostic terrain data.

Attributes

base_thickness_ratio

Ratio of the heightfield base thickness to its surface height.

border_width

Width of the flat outer border around the staircase, in meters.

flat_patch_sampling

Named flat-patch sampling configurations, or None to disable.

holes

If True, carve deep pits at the diagonal corners of each step ring.

horizontal_scale

Heightfield grid resolution.

pit_depth

Depth of holes-mode pits below the lowest stair, in meters.

platform_width

Side length of the flat square platform at the top of the staircase, in meters.

proportion

Robot spawning weight for this terrain type.

size

Width and length of the terrain patch, in meters.

vertical_scale

Heightfield height resolution.

step_height_range

Min and max step height, in meters.

step_width

Depth (run) of each step, in meters.
__init__(proportion=1.0, size=(10.0, 10.0), flat_patch_sampling=None, *, step_height_range, step_width, platform_width=1.0, border_width=0.0, holes=False, pit_depth=5.0, horizontal_scale=0.05, vertical_scale=0.005, base_thickness_ratio=1.0)
Parameters:

  • proportion (float)

  • size (tuple[float, float])

  • flat_patch_sampling (dict[str, FlatPatchSamplingCfg] | None)

  • step_height_range (tuple[float, float])

  • step_width (float)

  • platform_width (float)

  • border_width (float)

  • holes (bool)

  • pit_depth (float)

  • horizontal_scale (float)

  • vertical_scale (float)

  • base_thickness_ratio (float)

flat_patch_sampling: dict[str, FlatPatchSamplingCfg] | None = None

Named flat-patch sampling configurations, or None to disable.

proportion: float = 1.0

Robot spawning weight for this terrain type.

In curriculum mode, controls how many robots are spawned on this terrain’s column relative to other terrain types. Each terrain type always gets exactly one column; proportion only affects spawning distribution.

In random mode, controls the sampling probability for each patch.

size: tuple[float, float] = (10.0, 10.0)

Width and length of the terrain patch, in meters.

step_height_range: tuple[float, float]

Min and max step height, in meters. Interpolated by difficulty.

step_width: float

Depth (run) of each step, in meters. Must be a multiple of horizontal_scale.

platform_width: float = 1.0

Side length of the flat square platform at the top of the staircase, in meters.

border_width: float = 0.0

Width of the flat outer border around the staircase, in meters.

holes: bool = False

If True, carve deep pits at the diagonal corners of each step ring.

pit_depth: float = 5.0

Depth of holes-mode pits below the lowest stair, in meters.

horizontal_scale: float = 0.05

Heightfield grid resolution. Overwritten by TerrainGenerator.

vertical_scale: float = 0.005

Heightfield height resolution. Overwritten by TerrainGenerator.

base_thickness_ratio: float = 1.0

Ratio of the heightfield base thickness to its surface height.

function(difficulty, rng)[source]

Generate backend-agnostic terrain data.

Return type:

TerrainOutput

Returns:

TerrainOutput containing spawn origin and heightfield data.

Parameters: