Implementation:Haosulab ManiSkill LiftPegUpright

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

Overview

Concrete implementation of the peg lifting task environment in ManiSkill.

Description

The LiftPegUprightEnv requires a robot to move a peg lying flat on the table to an upright position. The peg is a two-colored cylinder (red and blue halves) with half-width 0.025m and half-length 0.12m.

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

Randomizations include the peg's xy position in the region [-0.1, 0.1]. Success requires the peg's y euler angle to be within 0.08 radians of pi/2 and the z position to be within 0.005m of the peg half-length (0.12m). The dense reward combines rotation alignment, z-position reward, and a reaching/grasping component.

Usage

Use this environment for testing orientation manipulation skills. The task requires grasping and reorienting an object from horizontal to vertical.

Code Reference

Source Location

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

Signature

@register_env("LiftPegUpright-v1", max_episode_steps=50)
class LiftPegUprightEnv(BaseEnv):
    SUPPORTED_ROBOTS = ["panda", "fetch"]
    agent: Union[Panda, Fetch]
    peg_half_width = 0.025
    peg_half_length = 0.12

Import

import gymnasium as gym
import mani_skill.envs
env = gym.make("LiftPegUpright-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, peg pose
reward	float	Dense reward from rotation alignment + z-position + reaching
terminated	bool	Whether episode ended by success
truncated	bool	Whether episode hit max steps (50)
info	dict	Contains success flag

Usage Examples

Basic Usage

import gymnasium as gym
import mani_skill.envs

env = gym.make("LiftPegUpright-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