Implementation:Haosulab ManiSkill TurnFaucet

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

Overview

Concrete implementation of the faucet turning task environment in ManiSkill using PartNet-Mobility faucet models.

Description

The TurnFaucetEnv requires a robot to turn a faucet handle by interacting with its articulated joint. Faucet models are sourced from the PartNet-Mobility dataset with varied geometries.

Registered as TurnFaucet-v1 with max_episode_steps=200 and asset_download_ids=["partnet_mobility_faucet"]. Supported robots: panda, panda_wristcam, fetch. Reward modes include "sparse" and "none".

The faucet is placed on a table and the robot must rotate its handle joint. Different faucet models have different handle geometries and joint configurations. The task tests the ability to manipulate articulated objects.

Usage

Use this environment for articulated object manipulation research. The variety of faucet geometries tests generalization across different handle shapes and joint types.

Code Reference

Source Location

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

Signature

@register_env(
    "TurnFaucet-v1",
    max_episode_steps=200,
    asset_download_ids=["partnet_mobility_faucet"],
)
class TurnFaucetEnv(BaseEnv):
    SUPPORTED_REWARD_MODES = ["sparse", "none"]
    SUPPORTED_ROBOTS = ["panda", "panda_wristcam", "fetch"]

Import

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

I/O Contract

Inputs

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

Outputs

Name	Type	Description
obs	dict/array	Observation including robot state, faucet joint angle, handle position
reward	float	Sparse reward based on faucet joint rotation progress
terminated	bool	Whether episode ended by success
truncated	bool	Whether episode hit max steps (200)
info	dict	Contains success flag

Usage Examples

Basic Usage

import gymnasium as gym
import mani_skill.envs

env = gym.make("TurnFaucet-v1", obs_mode="state", render_mode="rgb_array")
obs, info = env.reset()
for _ in range(200):
    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