Implementation:Haosulab ManiSkill EmptyEnv

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

Overview

Concrete implementation of an empty/dummy environment for showcasing robots in ManiSkill.

Description

The EmptyEnv is a minimal environment that places a robot on a ground plane with no objects or tasks. It is intended for testing robot loading, visualization, and basic interaction without any manipulation objectives.

Registered as Empty-v1 with max_episode_steps=200000. The default robot is panda, but any robot can be used. The reward mode is "none" only. The scene contains only a ground plane with collision enabled.

The evaluate method returns an empty dict, and no success/failure conditions exist.

Usage

Use this environment for testing robot configurations, visualizing robot models, debugging controller setups, or as a template for creating new environments.

Code Reference

Source Location

• Repository: Haosulab_ManiSkill
• File: mani_skill/envs/tasks/empty_env.py

Signature

@register_env("Empty-v1", max_episode_steps=200000)
class EmptyEnv(BaseEnv):
    SUPPORTED_REWARD_MODES = ["none"]
    def __init__(self, *args, robot_uids="panda", **kwargs): ...

Import

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

I/O Contract

Inputs

Name	Type	Required	Description
obs_mode	str	No	Observation mode
reward_mode	str	No	Reward mode: "none"
control_mode	str	No	Control mode for the robot
robot_uids	str	No	Robot to load (default: "panda")

Outputs

Name	Type	Description
obs	dict/array	Observation based on obs_mode (minimal)
reward	float	Always 0 (no reward)
terminated	bool	Always False
truncated	bool	Whether episode hit max steps (200000)
info	dict	Empty dict

Usage Examples

Basic Usage

import gymnasium as gym
import mani_skill.envs

env = gym.make("Empty-v1", obs_mode="state", render_mode="rgb_array")
obs, info = env.reset()
for _ in range(100):
    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