iGibson

https://stanfordvl.github.io/iGibson/
Admin
iGibson is a fast interactive simulation environment with physics and rendering for training and...
Tokens:72,743
Snippets:722
Trust Score:9.7
Update:5 months ago
Show doc for...
Tokens:
Raw
### Prepare ROS Environment for Running Examples

Source: https://stanfordvl.github.io/iGibson/ros_integration.html

Source the ROS environment and your catkin workspace's setup file to make ROS packages and your custom workspace available.

```bash
source /opt/ros/kinetic/setup.bash
source ~/catkin_ws/devel/setup.bash

```

--------------------------------

### Install Miniconda and Setup Environment

Source: https://stanfordvl.github.io/iGibson/_sources/installation.md.txt

Installs Miniconda, adds it to the PATH, updates conda, and creates a dedicated virtual environment for iGibson with Python 3.8.

```bash
curl -LO http://repo.continuum.io/miniconda/Miniconda-latest-Linux-x86_64.sh
bash Miniconda-latest-Linux-x86_64.sh
rm Miniconda-latest-Linux-x86_64.sh

echo "export PATH=$HOME/.miniconda/bin:$PATH" >> .bashrc 

conda update -y conda
conda create -y -n igibson python=3.8
conda activate igibson
```

--------------------------------

### Run Non-GUI Docker Example

Source: https://stanfordvl.github.io/iGibson/_sources/quickstart.md.txt

Launches a Docker container to run a non-GUI iGibson example. Ensure Docker is installed and the --gpus flag is used if GPU acceleration is needed.

```bash
docker run --gpus all -ti --rm igibson/igibson:latest /bin/bash
python -m tests.benchmark.benchmark_static_scene
```

--------------------------------

### Run Interactive Environment Example

Source: https://stanfordvl.github.io/iGibson/viewer.html

Execute this command to launch an interactive example of the iGibson Viewer. Ensure you have downloaded the Rs_int scene.

```bash
python -m igibson.examples.environments.env_int_example

```

--------------------------------

### Launch iGibson-ROS Examples

Source: https://stanfordvl.github.io/iGibson/_sources/ros_integration.md.txt

Launch various iGibson-ROS integration examples using roslaunch, including basic bringup, gmapping, and navigation.

```bash
roslaunch igibson-ros turtlebot_rgbd.launch # Bare minimal bringup example
roslaunch igibson-ros turtlebot_gmapping.launch # Run gmapping
roslaunch igibson-ros turtlebot_navigation.launch # Run the navigation stack, we have provided the map
```

--------------------------------

### Install Miniconda and Setup iGibson Environment on Linux

Source: https://stanfordvl.github.io/iGibson/installation.html

Installs Miniconda, updates conda, and creates a new virtual environment named 'igibson' with Python 3.8. This is recommended for managing iGibson dependencies.

```bash
# Install miniconda
curl -LO http://repo.continuum.io/miniconda/Miniconda-latest-Linux-x86_64.sh
bash Miniconda-latest-Linux-x86_64.sh
rm Miniconda-latest-Linux-x86_64.sh

# Add conda to your PATH
echo "export PATH=$HOME/.miniconda/bin:$PATH" >> .bashrc 

# Update conda and create a virtual environment for iGibson
conda update -y conda
conda create -y -n igibson python=3.8
conda activate igibson

```

--------------------------------

### Run Interactive Environment Example

Source: https://stanfordvl.github.io/iGibson/_sources/viewer.md.txt

Execute this command to launch an interactive environment example using the Viewer. Ensure you have downloaded the Rs_int scene.

```bash
python -m igibson.examples.environments.env_int_example
```

--------------------------------

### Install iGibson Simulator using Pip

Source: https://stanfordvl.github.io/iGibson/_sources/installation.md.txt

Installs the iGibson simulator as a Python package using pip. This is a quick method for users who do not need to modify the simulator's source code. It also includes a command to run a basic environment example.

```bash
pip install igibson  # This step takes about 4 minutes
# run the demo
python -m igibson.examples.environments.env_nonint_example
```

--------------------------------

### Run iGibson Robot Navigation Demo

Source: https://stanfordvl.github.io/iGibson/installation.html

Execute a simple robot navigation demo after installing the code and downloading demo data. This script provides a basic example of iGibson's capabilities.

```bash
python -m igibson.examples.environments.env_nonint_example

```

--------------------------------

### ROS Environment Setup

Source: https://stanfordvl.github.io/iGibson/_sources/ros_integration.md.txt

Source the necessary ROS setup files to configure your current shell environment for ROS communication.

```bash
source /opt/ros/kinetic/setup.bash
source ~/catkin_ws/devel/setup.bash
```

--------------------------------

### EGL Example Repository

Source: https://stanfordvl.github.io/iGibson/_sources/issues.md.txt

Refer to Erwin Coumans' egl_example repository for information on setting up OpenGL libraries and EGL.

```bash
git clone https://github.com/erwincoumans/egl_example.git
```

--------------------------------

### Visualize All Robots

Source: https://stanfordvl.github.io/iGibson/_sources/robots.md.txt

Run this example to quickly see all supported robots in iGibson.

```python
python -m igibson.examples.robots.all_robots_visualizer
```

--------------------------------

### Run Interactive Mesh Renderer Example

Source: https://stanfordvl.github.io/iGibson/renderer.html

Launches an interactive demo of the MeshRenderer. Use WASD to translate and mouse to rotate the camera. Press Q to exit.

```bash
python -m igibson.examples.renderer.mesh_renderer_example

```

--------------------------------

### Run GUI Docker Example (X11)

Source: https://stanfordvl.github.io/iGibson/_sources/quickstart.md.txt

Launches a Docker container with GUI access for X11 environments. This requires setting the DISPLAY environment variable and mounting the X11 socket.

```bash
docker run --env="DISPLAY" --volume="/tmp/.X11-unix:/tmp/.X11-unix:rw" --env="QT_X11_NO_MITSHM=1" --gpus all -ti --rm igibson/igibson
python -m igibson.examples.environments.env_nonint_example
```

--------------------------------

### Run Robot Control Example

Source: https://stanfordvl.github.io/iGibson/robots.html

Execute the example script to test robot control functionalities, including selecting robots, controllers, and input methods.

```python
python -m igibson.examples.robots.robot_control_example

```

--------------------------------

### Run Robot Control Example Script

Source: https://stanfordvl.github.io/iGibson/_sources/robots.md.txt

Execute the robot control example script to test different robots, controllers, and input methods.

```python
python -m igibson.examples.robots.robot_control_example
```

--------------------------------

### Install iGibson-ROS Dependencies

Source: https://stanfordvl.github.io/iGibson/ros_integration.html

Install the dependencies required for the iGibson-ROS package using rosdep.

```bash
cd ~/catkin_ws
rosdep install --from-paths src --ignore-src -r -y

```

--------------------------------

### Install iGibson from Source

Source: https://stanfordvl.github.io/iGibson/_sources/ros_integration.md.txt

Clone the iGibson repository and install it from source using pip. Ensure you are using Python 3.8 and have a conda environment activated.

```bash
git clone https://github.com/StanfordVL/iGibson --recursive
cd iGibson

# if you didn't create the conda environment before:
conda create -y -n igibson python=3.8
conda activate igibson

pip install -e .
```

--------------------------------

### Run Non-Interactive Environment Example

Source: https://stanfordvl.github.io/iGibson/_sources/quickstart.md.txt

Executes a non-interactive iGibson environment example. Reduce texture scaling in the config file if you encounter segfaults.

```bash
python -m igibson.examples.environments.env_nonint_example
```

--------------------------------

### Create and Populate Catkin Workspace

Source: https://stanfordvl.github.io/iGibson/_sources/ros_integration.md.txt

Set up a ROS catkin workspace, create the 'src' directory, and soft-link the 'igibson-ros' package into it. Then, build the workspace and install ROS dependencies.

```bash
mkdir -p ~/catkin_ws/src
cd ~/catkin_ws/src
ln -s $<iGibson root>/igibson/examples/ros/igibson-ros/ .

source /opt/ros/noetic/setup.bash # to setup ros environment
cd ~/catkin_ws && catkin_make

cd ~/catkin_ws
rosdep install --from-paths src --ignore-src -r -y
```

--------------------------------

### Launch iGibson-ROS Turtlebot Examples

Source: https://stanfordvl.github.io/iGibson/ros_integration.html

Launch various iGibson-ROS configurations for the Turtlebot, including a minimal bringup, gmapping, and navigation stack.

```bash
roslaunch igibson-ros turtlebot_rgbd.launch # Bare minimal bringup example
roslaunch igibson-ros turtlebot_gmapping.launch # Run gmapping
roslaunch igibson-ros turtlebot_navigation.launch # Run the navigation stack, we have provided the map

```

--------------------------------

### Run GUI Docker Example (VNC)

Source: https://stanfordvl.github.io/iGibson/_sources/quickstart.md.txt

Launches a Docker container with VNC support for headless environments. Access the GUI via a VNC client at <remote-ip>:5900.

```bash
docker run --gpus all -ti --rm igibson/igibson-vnc:latest /bin/bash
python -m igibson.examples.environments.env_nonint_example
```

--------------------------------

### Run PBR Mesh Renderer Example

Source: https://stanfordvl.github.io/iGibson/renderer.html

Tests the physically based renderer with a PBR demo. Requires a path to an object's visual folder in the iG dataset.

```bash
python -m igibson.examples.renderer.mesh_renderer_example_pbr <path to ig_dataset>/objects/sink/sink_1/shape/visual

```

--------------------------------

### Simple Mesh Rendering Example

Source: https://stanfordvl.github.io/iGibson/renderer.html

Demonstrates basic usage of MeshRenderer to load a scene, set up the camera, and render RGB, normal, and 3D images. Requires a scene model file.

```python
import logging
import os
import sys

import cv2
import numpy as np

from igibson.render.mesh_renderer.mesh_renderer_cpu import MeshRenderer
from igibson.utils.assets_utils import get_scene_path


def main(selection="user", headless=False, short_exec=False):
    """
    Minimal example of use of the renderer. Loads Rs (non interactive), renders one set of images (RGB, normals,
    3D points (as depth)), shows them.
    """
    print("*" * 80 + "\nDescription:" + main.__doc__ + "*" * 80)

    # If a model is given, we load it, otherwise we load Rs mesh (non interactive)
    if len(sys.argv) > 1:
        model_path = sys.argv[1]
    else:
        model_path = os.path.join(get_scene_path("Rs"), "mesh_z_up.obj")

    renderer = MeshRenderer(width=512, height=512)
    renderer.load_object(model_path)
    renderer.add_instance_group([0])
    camera_pose = np.array([0, 0, 1.2])
    view_direction = np.array([1, 0, 0])
    renderer.set_camera(camera_pose, camera_pose + view_direction, [0, 0, 1])
    renderer.set_fov(90)
    frames = renderer.render(modes=("rgb", "normal", "3d"))

    # Render 3d points as depth map
    depth = np.linalg.norm(frames[2][:, :, :3], axis=2)
    depth /= depth.max()
    frames[2][:, :, :3] = depth[..., None]

    if not headless:
        frames = cv2.cvtColor(np.concatenate(frames, axis=1), cv2.COLOR_RGB2BGR)
        cv2.imshow("image", frames)
        cv2.waitKey(0)


if __name__ == "__main__":
    logging.basicConfig(level=logging.INFO)
    main()

```

--------------------------------

### Compile iGibson from Source

Source: https://stanfordvl.github.io/iGibson/installation.html

Installs iGibson by cloning the GitHub repository and compiling from source. This method is recommended for developers who plan to modify the iGibson codebase.

```bash
git clone https://github.com/StanfordVL/iGibson --recursive
cd iGibson

# if you didn't create the conda environment before:
conda create -y -n igibson python=3.8
conda activate igibson

pip install -e . # This step takes about 4 minutes
```

--------------------------------

### Install iGibson from Source using Pip

Source: https://stanfordvl.github.io/iGibson/_sources/installation.md.txt

Installs iGibson in editable mode after cloning the repository and setting up the environment. This allows for direct modifications to the source code to be reflected immediately.

```bash
pip install -e . # This step takes about 4 minutes
```

--------------------------------

### Main Execution Block

Source: https://stanfordvl.github.io/iGibson/simulators.html

Configures logging and calls the main function to run the simulator examples.

```python
if __name__ == "__main__":
    logging.basicConfig(level=logging.INFO)
    main()
```

--------------------------------

### Install Dependencies for On-Screen Nvidia on Ubuntu 20.04

Source: https://stanfordvl.github.io/iGibson/_sources/installation.md.txt

Installs required Nvidia drivers and CUDA libraries for on-screen rendering on Ubuntu 20.04, suitable for desktop computers. This includes setting up Nvidia repositories and installing specific driver and CUDA versions.

```bash
# Add the nvidia ubuntu repositories
apt-get update && apt-get install -y --no-install-recommends \
    gnupg2 curl ca-certificates && \
    curl -fsSL https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2004/x86_64/7fa2af80.pub | apt-key add - && \
    echo "deb https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2004/x86_64 /" > /etc/apt/sources.list.d/cuda.list && \
    echo "deb https://developer.download.nvidia.com/compute/machine-learning/repos/ubuntu2004/x86_64 /" > /etc/apt/sources.list.d/nvidia-ml.list

# The following cuda libraries are required to compile igibson
apt-get update && apt-get update && apt-get install -y --no-install-recommends \
    xserver-xorg-video-nvidia-470 \
    cuda-cudart-11-1=11.1.74-1 \
    cuda-compat-11-1 \
    cuda-command-line-tools-11-1=11.1.1-1 \
    cuda-libraries-dev-11-1=11.1.1-1 \

```

--------------------------------

### Test iGibson Installation

Source: https://stanfordvl.github.io/iGibson/_sources/installation.md.txt

Verify your iGibson installation by importing the library in a Python interpreter. This is a basic check to ensure the package is accessible.

```python
import igibson
```

--------------------------------

### Load Objects into iGibson Environment

Source: https://stanfordvl.github.io/iGibson/objects.html

This example demonstrates importing objects into different types of iGibson scenes (empty, iG, static). It shows how to initialize the simulator, scene, and robot, and then load objects with specified positions and orientations. Use this to set up a scene with specific objects and a robot for interaction.

```python
import logging
import os
from sys import platform

import numpy as np
import yaml

import igibson
from igibson.envs.igibson_env import iGibsonEnv
from igibson.external.pybullet_tools.utils import quat_from_euler
from igibson.objects.articulated_object import URDFObject
from igibson.objects.ycb_object import YCBObject
from igibson.render.mesh_renderer.mesh_renderer_cpu import MeshRendererSettings
from igibson.render.profiler import Profiler
from igibson.robots.turtlebot import Turtlebot
from igibson.scenes.empty_scene import EmptyScene
from igibson.scenes.gibson_indoor_scene import StaticIndoorScene
from igibson.simulator import Simulator
from igibson.utils.assets_utils import get_ig_avg_category_specs, get_ig_category_path, get_ig_model_path
from igibson.utils.utils import let_user_pick, parse_config


def main(selection="user", headless=False, short_exec=False):
    """
    This demo shows how to load scaled objects from the iG object model dataset and
    additional objects from the YCB dataset in predefined locations
    Loads a concrete object model of a table, and a random one of the same category, and 10 cracker boxes
    The objects can be loaded into an empty scene, an interactive scene (iG) or a static scene (Gibson)
    The example also shows how to use the Environment API or directly the Simulator API, loading objects and robots
    and executing actions
    """
    print("*" * 80 + "\nDescription:" + main.__doc__ + "*" * 80)
    scene_options = get_first_options()
    type_of_scene = let_user_pick(scene_options, selection=selection) - 1

    if type_of_scene == 0:  # Empty
        config = parse_config(os.path.join(igibson.configs_path, "turtlebot_static_nav.yaml"))
        settings = MeshRendererSettings(enable_shadow=False, msaa=False, texture_scale=0.5)
        s = Simulator(
            mode="gui_interactive" if not headless else "headless",
            image_width=512,
            image_height=512,
            rendering_settings=settings,
        )
        scene = EmptyScene(floor_plane_rgba=[0.6, 0.6, 0.6, 1])
        # scene.load_object_categories(benchmark_names)
        s.import_scene(scene)
        robot_config = config["robot"]
        robot_config.pop("name")
        turtlebot = Turtlebot(**robot_config)
        s.import_object(turtlebot)

    elif type_of_scene == 1:  # iG
        config_filename = os.path.join(igibson.configs_path, "turtlebot_nav.yaml")
        config_data = yaml.load(open(config_filename, "r"), Loader=yaml.FullLoader)
        config_data["load_object_categories"] = []  # Uncomment this line to accelerate loading with only the building
        config_data["visible_target"] = False
        config_data["visible_path"] = False
        # Reduce texture scale for Mac.
        if platform == "darwin":
            config_data["texture_scale"] = 0.5
        env = iGibsonEnv(config_file=config_data, mode="gui_interactive" if not headless else "headless")
        s = env.simulator

    elif type_of_scene == 2:  # Gibson
        config = parse_config(os.path.join(igibson.configs_path, "turtlebot_static_nav.yaml"))
        settings = MeshRendererSettings(enable_shadow=False, msaa=False)
        # Reduce texture scale for Mac.
        if platform == "darwin":
            settings.texture_scale = 0.5
        s = Simulator(
            mode="gui_interactive" if not headless else "headless",
            image_width=512,
            image_height=512,
            rendering_settings=settings,
        )

        scene = StaticIndoorScene("Rs", build_graph=True, pybullet_load_texture=False)
        s.import_scene(scene)
        robot_config = config["robot"]
        robot_config.pop("name")
        turtlebot = Turtlebot(**robot_config)
        s.import_object(turtlebot)

    if not headless:
        # Set a better viewing direction
        s.viewer.initial_pos = [-2, 1.4, 1.2]
        s.viewer.initial_view_direction = [0.6, -0.8, 0.1]
        s.viewer.reset_viewer()

    # Objects to load: two tables, the first one is predefined model, the second, random for the same category
    table_objects_to_load = {
        "table_1": {
            "category": "breakfast_table",
            "model": "1b4e6f9dd22a8c628ef9d976af675b86",
            "pos": (0.0, -0.2, 1.01),
            "orn": (0, 0, 90),
        },
        "table_2": {
            "category": "breakfast_table",
            "pos": (0.5, -2.0, 1.01),
            "orn": (0, 0, 45),
        },
    }

    # Load the specs of the object categories, e.g., common scaling factor
    avg_category_spec = get_ig_avg_category_specs()

    scene_objects = {}
    try:
        for obj in table_objects_to_load.values():
            category = obj["category"]
            if category in scene_objects:
                scene_objects[category] += 1
            else:
                scene_objects[category] = 1

```

--------------------------------

### Install System Dependencies for iGibson

Source: https://stanfordvl.github.io/iGibson/_sources/installation.md.txt

Installs essential system packages required for building iGibson. Ensure CMake, Git, G++, and EGL development libraries are present.

```bash
apt-get update && apt-get install -y --no-install-recommends \
    cmake \
    git \
    g++ \
    libegl-dev
```

--------------------------------

### Install Linux Dependencies for Headless iGibson (e.g., Cluster)

Source: https://stanfordvl.github.io/iGibson/installation.html

Installs necessary system dependencies for iGibson on Ubuntu 20.04, including Nvidia drivers and CUDA libraries for headless operation. Run as root/superuser.

```bash
# Add the nvidia ubuntu repositories
apt-get update && apt-get install -y --no-install-recommends \
    gnupg2 curl ca-certificates && \
    curl -fsSL https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2004/x86_64/7fa2af80.pub | apt-key add - && \
    echo "deb https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2004/x86_64 /" > /etc/apt/sources.list.d/cuda.list && \
    echo "deb https://developer.download.nvidia.com/compute/machine-learning/repos/ubuntu2004/x86_64 /" > /etc/apt/sources.list.d/nvidia-ml.list

# The following cuda libraries are required to compile igibson
apt-get update && apt-get update && apt-get install -y --no-install-recommends \
    nvidia-headless-470 \
    cuda-cudart-11-1=11.1.74-1 \
    cuda-compat-11-1 \
    cuda-command-line-tools-11-1=11.1.1-1 \
    cuda-libraries-dev-11-1=11.1.1-1 \

# For building and running igibson
apt-get update && apt-get install -y --no-install-recommends \
    cmake \
    git \
    g++ \
    libegl-dev

```

--------------------------------

### Instantiate and Step iGibsonEnv with Static Scene

Source: https://stanfordvl.github.io/iGibson/_sources/environments.md.txt

Use this snippet to create an iGibsonEnv with a static scene and step through episodes. Remember to call env.reset() at the start of each episode and env.step(action) to advance.

```python
import pybullet
import pybullet_data

import igibson
from igibson.envs import iGibsonEnv

# Define the config file
config_file = "igibson/configs/turtlebot_static_nav.yaml"

# Create the environment
env = iGibsonEnv(config_file=config_file)

# Reset the environment at the beginning of each episode
state, _ = env.reset()

# Step through the environment
for _ in range(1000):
    # Sample a random action
    action = env.action_space.sample()
    # Step the environment
    state, reward, done, info = env.step(action)
    # Check if the episode is done
    if done:
        print("Episode finished.")
        state, _ = env.reset()

env.close()

```

--------------------------------

### Load and Interact with iGibson Indoor Scenes

Source: https://stanfordvl.github.io/iGibson/scenes.html

Imports an interactive scene and demonstrates sampling points by room type and computing shortest paths. Useful for tasks requiring robots to spawn in specific room types. Requires iGibson dataset and simulator setup.

```python
import logging
from sys import platform

import numpy as np

from igibson.render.mesh_renderer.mesh_renderer_settings import MeshRendererSettings
from igibson.render.profiler import Profiler
from igibson.scenes.igibson_indoor_scene import InteractiveIndoorScene
from igibson.simulator import Simulator
from igibson.utils.assets_utils import get_available_ig_scenes
from igibson.utils.utils import let_user_pick


def main(selection="user", headless=False, short_exec=False):
    """
    Prompts the user to select any available interactive scene and loads it.
    Shows how to load directly scenes without the Environment interface
    Shows how to sample points in the scene by room type and how to compute geodesic distance and the shortest path
    """
    print("*" * 80 + "\nDescription:" + main.__doc__ + "*" * 80)
    available_ig_scenes = get_first_options()
    scene_id = available_ig_scenes[let_user_pick(available_ig_scenes, selection=selection) - 1]
    settings = MeshRendererSettings(enable_shadow=True, msaa=False)
    if platform == "darwin":
        settings.texture_scale = 0.5
    s = Simulator(
        mode="gui_interactive" if not headless else "headless",
        image_width=512,
        image_height=512,
        rendering_settings=settings,
    )

    scene = InteractiveIndoorScene(
        scene_id,
        # load_object_categories=[],  # To load only the building. Fast
        build_graph=True,
    )
    s.import_scene(scene)

    # Shows how to sample points in the scene
    np.random.seed(0)
    for _ in range(10):
        pt = scene.get_random_point_by_room_type("living_room")[1]
        print("Random point in living_room: {}".format(pt))

    for _ in range(10):
        random_floor = scene.get_random_floor()
        p1 = scene.get_random_point(random_floor)[1]
        p2 = scene.get_random_point(random_floor)[1]
        shortest_path, geodesic_distance = scene.get_shortest_path(random_floor, p1[:2], p2[:2], entire_path=True)
        print("Random point 1: {}".format(p1))
        print("Random point 2: {}".format(p2))
        print("Geodesic distance between p1 and p2: {}".format(geodesic_distance))
        print("Shortest path from p1 to p2: {}".format(shortest_path))

    if not headless:
        input("Press enter")

    max_steps = -1 if not short_exec else 1000
    step = 0
    while step != max_steps:
        with Profiler("Simulator step"):
            s.step()
            step += 1

    s.disconnect()


def get_first_options():
    return get_available_ig_scenes()


if __name__ == "__main__":
    logging.basicConfig(level=logging.INFO)
    main()

```

--------------------------------

### Install iGibson via Pip

Source: https://stanfordvl.github.io/iGibson/installation.html

Installs the iGibson simulator as a Python package using pip. This is a quick method for users who do not plan to modify the simulator's source code.

```bash
pip install igibson  # This step takes about 4 minutes
```

```python
python -m igibson.examples.environments.env_nonint_example
```

--------------------------------

### Import Objects into iGibson

Source: https://stanfordvl.github.io/iGibson/_sources/objects.md.txt

This example demonstrates how to import various objects into the iGibson simulator. Ensure that the necessary object assets are available in the specified paths.

```python
import pybullet as p

import igibson
from igibson.objects.ycb import YCBObject
from igibson.objects.rbo import RBOObject
from igibson.objects.shape import ShapeNetObject
from igibson.objects.articulated_object import URDFObject
from igibson.objects.visual_marker import VisualMarker
from igibson.simulator import Simulator
from igibson.utils.utils import parse_config

def main():
    # Initialize the simulator
    config = parse_config(os.path.join(igibson.root_path, 'configs', 'config.yaml'))
    env = Simulator(config=config)

    # Load YCB objects
    ycb_obj_path = os.path.join(igibson.assets_path, 'models', 'ycb')
    obj_name = '003_cracker_box'
    ycb_obj = YCBObject(os.path.join(ycb_obj_path, obj_name), name=obj_name)
    env.import_object(ycb_obj, use_aabb_center=True)

    # Load RBO objects
    rbo_obj_path = os.path.join(igibson.assets_path, 'models', 'rbo')
    obj_name = '0001_sugar_box'
    rbo_obj = RBOObject(os.path.join(rbo_obj_path, obj_name), name=obj_name)
    env.import_object(rbo_obj, use_aabb_center=True)

    # Load ShapeNet objects
    shapenet_obj_path = os.path.join(igibson.assets_path, 'models', 'shapenet')
    obj_name = '037_monitor'
    shapenet_obj = ShapeNetObject(os.path.join(shapenet_obj_path, obj_name), name=obj_name)
    env.import_object(shapenet_obj, use_aabb_center=True)

    # Load URDF objects
    urdf_obj_path = os.path.join(igibson.assets_path, 'models', 'urdf')
    obj_name = '0001_sugar_box'
    urdf_obj = URDFObject(os.path.join(urdf_obj_path, obj_name), name=obj_name)
    env.import_object(urdf_obj, use_aabb_center=True)

    # Load Visual Marker
    marker_path = os.path.join(igibson.assets_path, 'models', 'marker')
    marker_obj = VisualMarker(os.path.join(marker_path, 'marker.urdf'), name='marker')
    env.import_object(marker_obj, use_aabb_center=True)

    # Run simulation for a few steps
    for _ in range(1000):
        env.step()

    env.disconnect()

if __name__ == "__main__":
    main()

```

--------------------------------

### Simple Mesh Rendering Example

Source: https://stanfordvl.github.io/iGibson/_sources/renderer.md.txt

Renders an iGibson scene using the MeshRenderer. Ensure the scene and renderer are properly initialized before calling render.

```python
import os
import numpy as np
import pybullet as p

import igibson
from igibson.scenes import Scene
from igibson.objects.ycb.ycb_object import YCBObject
from igibson.utils.utils import quat2rotmat, xyzw2wxyz
from igibson.render.mesh_renderer import MeshRenderer

def mesh_renderer_simple_example():
    # Initialize the renderer
    # You can specify the image width, height, and vertical field of view
    renderer = MeshRenderer(width=960, height=720, vertical_fov=90)

    # Load a scene
    scene_name = "Rs"
    scene = Scene(os.path.join(igibson.ig_dataset_path, "scenes", scene_name))
    env = igibson.simulator.Simulator(scene)

    # Add the scene to the renderer
    renderer.add_scene(scene)

    # Set the camera pose
    # The camera pose is defined by position and orientation (quaternion)
    # Here, we set the camera to look at the center of the scene
    camera_pose = env.get_camera_pose()
    renderer.set_camera(camera_pose[0], camera_pose[1])

    # Render the scene in different modes
    # Supported modes: 'rgb', 'normal', 'seg', '3d', 'optical_flow', 'scene_flow'
    # The output is a dictionary of numpy arrays, where keys are the modes
    rendering_output = renderer.render(modes=("rgb", "normal", "seg", "3d", "optical_flow", "scene_flow"))

    # You can access the rendered images like this:
    # rgb_image = rendering_output["rgb"]
    # normal_image = rendering_output["normal"]
    # seg_image = rendering_output["seg"]
    # depth_image = rendering_output["3d"][:, :, 2]
    # optical_flow_image = rendering_output["optical_flow"]
    # scene_flow_image = rendering_output["scene_flow"]

    # Clean up the renderer and environment
    renderer.delete()
    env.disconnect()

if __name__ == "__main__":
    mesh_renderer_simple_example()

```

--------------------------------

### Load Objects into Simulator - Python

Source: https://stanfordvl.github.io/iGibson/_sources/simulators.md.txt

This example demonstrates how to import a scene, a Turtlebot, and multiple YCBObjects into the simulator. Ensure the necessary scene and object types are available.

```python
import pybullet as p
import numpy as np

from igibson.envs.env_base import BaseEnv
from igibson.scenes import InteractiveIndoorScene
from igibson.objects.ycb import YCBObject
from igibson.robots import Turtlebot
from igibson.utils.utils import random_string


class LoadObjectsEnv(BaseEnv):
    def __init__(self, config):
        super(LoadObjectsEnv, self).__init__(config)

    def _load_scene(self):
        scene_type = self.config.get("scene_type", "Rs")
        if scene_type == "Rs":
            scene = InteractiveIndoorScene(scene_id=scene_type)
        else:
            raise NotImplementedError
        return scene

    def _load_object(self, obj_name, placement):
        obj = YCBObject(obj_name)
        self.simulator.import_object(obj, placement=placement)
        return obj

    def _load_articulated_object(self, ao_name, placement):
        ao = YCBObject(ao_name)
        self.simulator.import_articulated_object(ao, placement=placement)
        return ao

    def _load_robot(self, robot_name, placement):
        if robot_name == "turtlebot":
            robot = Turtlebot(self.config)
            self.simulator.import_robot(robot, placement=placement)
            return robot
        else:
            raise NotImplementedError

    def reset(self):
        self.simulator.reload()
        self.scene = self._load_scene()
        self.simulator.import_scene(self.scene)

        # Load objects
        obj_name = "coffee_mug"
        obj_placement = np.array([0.5, 0.5, 0.8])
        self.obj = self._load_object(obj_name, obj_placement)

        # Load articulated objects
        ao_name = "cheezit_box"
        ao_placement = np.array([0.6, 0.5, 0.8])
        self.ao = self._load_articulated_object(ao_name, ao_placement)

        # Load robot
        robot_name = "turtlebot"
        robot_placement = np.array([0.7, 0.5, 0.8])
        self.robot = self._load_robot(robot_name, robot_placement)

        # Load multiple YCB objects
        for i in range(10):
            obj_name = np.random.choice(YCBObject.get_all_object_names())
            obj_scale = np.random.uniform(0.5, 1.0)
            obj_orientation = p.getQuaternionFromEuler(np.random.uniform(0, np.pi, 3))
            obj_pose = np.array([np.random.uniform(-0.5, 0.5), np.random.uniform(-0.5, 0.5), 0.8])
            obj = YCBObject(obj_name, scale=obj_scale)
            self.simulator.import_object(obj, pose=obj_pose, orientation=obj_orientation)

        return self.get_observation()

    def get_observation(self):
        return np.array([])

    def step(self, action):
        # No-op for this example
        return self.get_observation(), 0.0, False, {}

    def close(self):
        self.simulator.disconnect()


def main():
    config = {
        "scene_type": "Rs",
        "render_options": {
            "viewer_kwargs": {
                "image_width": 512,
                "image_height": 512
            }
        }
    }
    env = LoadObjectsEnv(config)
    env.reset()
    # Simulate for a few steps
    for _ in range(100):
        action = env.action_space.sample()  # Replace with actual actions
        obs, reward, done, info = env.step(action)
        if done:
            break
    env.close()

if __name__ == "__main__":
    main()

```

--------------------------------

### Check NVIDIA Driver and EGL Installation

Source: https://stanfordvl.github.io/iGibson/_sources/issues.md.txt

Verify NVIDIA driver version and libegl1 installation. Ensure OpenGL libraries are visible and EGL is linked to Nvidia's implementation, not Mesa.

```bash
nvidia-smi
```

```bash
apt list --installed | grep libegl1
```

```bash
export LD_LIBRARY_PATH=/usr/lib/nvidia-<vvv>:$LD_LIBRARY_PATH
```

```bash
ldconfig -p | grep EGL
```

--------------------------------

### Docker Environment Setup for iGibson-ROS

Source: https://stanfordvl.github.io/iGibson/_sources/ros_integration.md.txt

Clone the iGibson repository and use the provided scripts to build and run the ROS environment within a Docker container.

```bash
git clone https://github.com/StanfordVL/iGibson --recursive

cd iGibson/docker/igibson-ros

./build.sh
./run_gui.sh
```

--------------------------------

### Build and Run iGibson-ROS Docker Image

Source: https://stanfordvl.github.io/iGibson/ros_integration.html

Clone the iGibson repository and use the provided Dockerfile to build and run the iGibson-ROS environment. This simplifies setup by containerizing dependencies.

```bash
git clone https://github.com/StanfordVL/iGibson --recursive

cd iGibson/docker/igibson-ros

./build.sh
./run_gui.sh

```

--------------------------------

### Get Scene Options

Source: https://stanfordvl.github.io/iGibson/objects.html

Returns a list of available scene types for selection.

```python
return ["Empty scene", "Interactive scene (iG)", "Static scene (Gibson)"]

```

--------------------------------

### Example EGL Library Linking Output

Source: https://stanfordvl.github.io/iGibson/issues.html

This output shows typical EGL library links. Verify that libEGL_nvidia is present and correctly linked, and avoid links to mesa-egl.

```text
libEGL_nvidia.so.0 (libc6,x86-64) => /usr/lib/x86_64-linux-gnu/libEGL_nvidia.so.0
	libEGL_nvidia.so.0 (libc6) => /usr/lib/i386-linux-gnu/libEGL_nvidia.so.0
	libEGL_mesa.so.0 (libc6,x86-64) => /usr/lib/x86_64-linux-gnu/libEGL_mesa.so.0
	libEGL.so.1 (libc6,x86-64) => /usr/lib/x86_64-linux-gnu/libEGL.so.1
	libEGL.so.1 (libc6) => /usr/lib/i386-linux-gnu/libEGL.so.1
	libEGL.so (libc6,x86-64) => /usr/lib/x86_64-linux-gnu/libEGL.so
	libEGL.so (libc6) => /usr/lib/i386-linux-gnu/libEGL.so
```

--------------------------------

### Load Static Scene and Compute Shortest Path

Source: https://stanfordvl.github.io/iGibson/_sources/scenes.md.txt

Loads a static indoor scene, samples random start and end locations, and computes the shortest path between them using the scene's traversability graph. Requires `igibson.g_dataset_path` to be set.

```python
import igibson

# Define the scene ID for a static indoor scene
scene_id = 'Rs_train'

# Load the static indoor scene
scene = igibson.scenes.StaticIndoorScene(scene_id=scene_id)

# Load the scene into the simulator
simulator.import_scene(scene)

# Sample a random start location
start_pos = scene.get_random_point(nav_mesh_path=scene.get_nav_mesh_path(0))

# Sample a random end location
end_pos = scene.get_random_point(nav_mesh_path=scene.get_nav_mesh_path(0))

# Compute the shortest path between the start and end locations
path = scene.get_shortest_path(start_pos, end_pos, scene.get_nav_mesh_path(0))

# Print the computed path (list of waypoints)
print(path)

```

--------------------------------

### run_setup

Source: https://stanfordvl.github.io/iGibson/apidoc/igibson.examples.web_ui.html

Sets up the environment for sampling. Can specify a list of scenes to load.

```APIDOC
## run_setup

### Description
Sets up the environment for sampling.

### Parameters
- **scenes** (list, optional) - A list of scenes to load.
```

--------------------------------

### Clone iGibson and Setup Conda Environment

Source: https://stanfordvl.github.io/iGibson/ros_integration.html

Clone the iGibson repository and set up a dedicated Conda environment for Python 3.8. This is a prerequisite for installing iGibson from source.

```bash
git clone https://github.com/StanfordVL/iGibson --recursive
cd iGibson

# if you didn't create the conda environment before:
conda create -y -n igibson python=3.8
conda activate igibson

pip install -e . # This step takes about 4 minutes

```

--------------------------------

### Install CMake on MacOS

Source: https://stanfordvl.github.io/iGibson/_sources/issues.md.txt

Use Homebrew to install CMake system-wide to resolve 'ModuleNotFoundError: No module named 'cmake''. Avoid using pip or conda installations for CMake.

```bash
brew install cmake
```

--------------------------------

### Install SVL PyBullet Fork

Source: https://stanfordvl.github.io/iGibson/_sources/installation.md.txt

Uninstalls the standard PyBullet package and installs the custom SVL fork required by iGibson. This is crucial for compatibility and performance, especially if you have a pre-existing PyBullet installation.

```bash
pip uninstall pybullet
pip install pybullet-svl
```

--------------------------------

### Install freetype on arm64 MacOS

Source: https://stanfordvl.github.io/iGibson/_sources/issues.md.txt

Address 'RuntimeError: Freetype library not found' on arm64 MacOS by installing freetype using conda, as the pip package may install incorrect binaries.

```bash
conda install freetype
```

--------------------------------

### Check Installed libegl1

Source: https://stanfordvl.github.io/iGibson/issues.html

Use this command to verify if the libegl1 package is installed on your system. This is a prerequisite for EGL functionality.

```bash
apt list --installed | grep libegl1
```

--------------------------------

### Install Dependencies for Headless Nvidia on Ubuntu 20.04

Source: https://stanfordvl.github.io/iGibson/_sources/installation.md.txt

Installs required Nvidia drivers and CUDA libraries for headless operation on Ubuntu 20.04, suitable for cluster environments. This includes setting up Nvidia repositories and installing specific driver and CUDA versions.

```bash
# Add the nvidia ubuntu repositories
apt-get update && apt-get install -y --no-install-recommends \
    gnupg2 curl ca-certificates && \
    curl -fsSL https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2004/x86_64/7fa2af80.pub | apt-key add - && \
    echo "deb https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2004/x86_64 /" > /etc/apt/sources.list.d/cuda.list && \
    echo "deb https://developer.download.nvidia.com/compute/machine-learning/repos/ubuntu2004/x86_64 /" > /etc/apt/sources.list.d/nvidia-ml.list

# The following cuda libraries are required to compile igibson
apt-get update && apt-get update && apt-get install -y --no-install-recommends \
    nvidia-headless-470 \
    cuda-cudart-11-1=11.1.74-1 \
    cuda-compat-11-1 \
    cuda-command-line-tools-11-1=11.1.1-1 \
    cuda-libraries-dev-11-1=11.1.1-1 \

# For building and running igibson
apt-get update && apt-get install -y --no-install-recommends \
    cmake \
    git \
    g++ \
    libegl-dev
```

--------------------------------

### Run Interactive Mesh Renderer Demo

Source: https://stanfordvl.github.io/iGibson/_sources/renderer.md.txt

Launches an interactive demo of the MeshRenderer. Use WASD to translate the camera and mouse to rotate. Press Q to exit.

```bash
python -m igibson.examples.renderer.mesh_renderer_example
```

--------------------------------

### Install h5py on arm64 MacOS

Source: https://stanfordvl.github.io/iGibson/_sources/issues.md.txt

Resolve 'ERROR: Could not build wheels for h5py' on arm64 MacOS by installing h5py using conda.

```bash
conda install h5py
```

--------------------------------

### Instantiate and Step iGibsonEnv with Static Scene

Source: https://stanfordvl.github.io/iGibson/environments.html

Instantiates an iGibsonEnv with a static Gibson model and demonstrates stepping through the environment with random actions. Requires downloading assets and demo data. Note: Mac users may need to reduce texture scaling to avoid memory errors.

```python
import logging
import os
from sys import platform

import yaml

import igibson
from igibson.envs.igibson_env import iGibsonEnv
from igibson.render.profiler import Profiler
from igibson.utils.assets_utils import download_assets, download_demo_data


def main(selection="user", headless=False, short_exec=False):
    """
    Creates an iGibson environment from a config file with a turtlebot in Rs (not interactive).
    It steps the environment 100 times with random actions sampled from the action space,
    using the Gym interface, resetting it 10 times.
    """
    print("*" * 80 + "\nDescription:" + main.__doc__ + "*" * 80)
    # If they have not been downloaded before, download assets and Rs Gibson (non-interactive) models
    download_assets()
    download_demo_data()
    config_filename = os.path.join(igibson.configs_path, "turtlebot_static_nav.yaml")
    config_data = yaml.load(open(config_filename, "r"), Loader=yaml.FullLoader)
    # Reduce texture scale for Mac.
    if platform == "darwin":
        config_data["texture_scale"] = 0.5

    # Shadows and PBR do not make much sense for a Gibson static mesh
    config_data["enable_shadow"] = False
    config_data["enable_pbr"] = False

    env = iGibsonEnv(config_file=config_data, mode="gui_interactive" if not headless else "headless")
    max_iterations = 10 if not short_exec else 1
    for j in range(max_iterations):
        print("Resetting environment")
        env.reset()
        for i in range(100):
            with Profiler("Environment action step"):
                action = env.action_space.sample()
                state, reward, done, info = env.step(action)
                if done:
                    print("Episode finished after {} timesteps".format(i + 1))
                    break
    env.close()


if __name__ == "__main__":
    logging.basicConfig(level=logging.INFO)
    main()

```

--------------------------------

### Verify CUDA Compiler (nvcc) Installation

Source: https://stanfordvl.github.io/iGibson/_sources/installation.md.txt

Checks if the NVIDIA CUDA compiler (nvcc) is accessible and displays its version. This confirms that your CUDA installation and PATH configuration are correct.

```bash
$ nvcc --version
nvcc: NVIDIA (R) Cuda compiler driver
Copyright (c) 2005-2020 NVIDIA Corporation
Built on Mon_Oct_12_20:09:46_PDT_2020
Cuda compilation tools, release 11.1, V11.1.105
Build cuda_11.1.TC455_06.29190527_0
```

--------------------------------

### Download Dataset using URL

Source: https://stanfordvl.github.io/iGibson/_sources/dataset.md.txt

Use this command to download, decompress, and place the dataset in the correct folder. Replace URL with the provided download link after filling out the license agreement.

```bash
python -m igibson.utils.assets_utils --download_dataset URL
```

--------------------------------

### Run PBR Renderer Demo

Source: https://stanfordvl.github.io/iGibson/_sources/renderer.md.txt

Tests the physically based renderer (PBR) with a specified object folder. Ensure the path points to a directory containing .obj files.

```bash
python -m igibson.examples.renderer.mesh_renderer_example_pbr <path to ig_dataset>/objects/sink/sink_1/shape/visual
```

--------------------------------

### Identity Matrix

Source: https://stanfordvl.github.io/iGibson/apidoc/igibson.external.pybullet_tools.html

Returns a 4x4 identity matrix. Useful as a starting point for transformations or for verification.

```python
>>> I = identity_matrix()
>>> numpy.allclose(I, numpy.dot(I, I))
True
>>> numpy.sum(I), numpy.trace(I)
(4.0, 4.0)
>>> numpy.allclose(I, numpy.identity(4, dtype=numpy.float64))
True
```

--------------------------------

### Render to PyTorch Tensors

Source: https://stanfordvl.github.io/iGibson/renderer.html

Shows how MeshRenderer can directly render into PyTorch tensors for increased efficiency. Requires PyTorch installation.

```python
# Code for rendering to PyTorch Tensors is not provided in the source text.
# The description indicates it's in igibson/examples/renderer/mesh_renderer_gpu_example.py

```
Tokens:
Raw