Implementation:Haosulab ManiSkill StackCube

Knowledge Sources	Haosulab_ManiSkill
Domains	Robotics, Simulation, Tabletop_Manipulation
Last Updated	2026-02-15 08:00 GMT

Overview

Concrete implementation of the cube stacking task environment in ManiSkill.

Description

The StackCubeEnv requires a robot to pick up a red cube and stack it on top of a green cube, then release it without the red cube falling off.

Registered as StackCube-v1 with max_episode_steps=50. Supported robots: panda and fetch. Reward modes include "normalized_dense", "dense", "sparse", and "none".

Randomizations: both cubes have z-axis rotation randomized, xy positions randomized on the table such that they do not collide. Success requires the red cube to be on top of the green cube (within half cube size), the red cube to be static, and the robot to not be grasping the red cube.

Usage

Use this environment for sequential manipulation research. Stacking requires grasping, precise placement, and careful release -- a fundamental skill in robotic manipulation.

Code Reference

Source Location

• Repository: Haosulab_ManiSkill
• File: mani_skill/envs/tasks/tabletop/stack_cube.py

Signature

@register_env("StackCube-v1", max_episode_steps=50)
class StackCubeEnv(BaseEnv):
    SUPPORTED_ROBOTS = ["panda", "fetch"]
    agent: Union[Panda, Fetch]

Import

import gymnasium as gym
import mani_skill.envs
env = gym.make("StackCube-v1")

I/O Contract

Inputs

Name	Type	Required	Description
obs_mode	str	No	Observation mode
reward_mode	str	No	Reward mode: "normalized_dense", "dense", "sparse", "none"
control_mode	str	No	Control mode for the robot

Outputs

Name	Type	Description
obs	dict/array	Observation including TCP pose, cubeA/cubeB poses, is_grasped
reward	float	Dense reward: reaching + grasping + placing + static + release
terminated	bool	Whether episode ended by success
truncated	bool	Whether episode hit max steps (50)
info	dict	Contains success, is_cubeA_on_cubeB, is_cubeA_static, is_grasped

Usage Examples

Basic Usage

import gymnasium as gym
import mani_skill.envs

env = gym.make("StackCube-v1", obs_mode="state", render_mode="rgb_array")
obs, info = env.reset()
for _ in range(50):
    action = env.action_space.sample()
    obs, reward, terminated, truncated, info = env.step(action)
    if terminated or truncated:
        obs, info = env.reset()
env.close()

Related Pages

• Environment:Haosulab_ManiSkill_Python_SAPIEN_Core