Ft easy adding walls issue: #77

docs/source/adding_to_envs.rst

@@ -0,0 +1,82 @@
+Adding elements to the environment
+===================================
+
+Walls
+-----
+
+To add a wall use:
+
+.. code:: python
+
+    env.add_walls(dim, poses_2d)
+
+
+dim: numpy.ndarray or list
+| (optional) dim, short for dimensions if given dim must
+| be of length 3 corresponding to [width, length, height].
+
+| poses_2d: [np.ndarray, ...] or [list, ...]
+| (optional) if given poses_2d must be a list with lists or numpy arrays of length 3
+| correspond to [x_position, y_position, orientation] which refers to walls centers of mass.
+| The height of the center of mass will be adjusted such that the bottom of the wall
+| aligns with the ground plane. multiple walls can be placed
+| by having multiple pose_2d in poses_2d.
+
+Shapes
+-------
+
+To add a shape use:
+
+.. code:: python
+
+    env.add_shapes(shape_type, dim, mass, poses_2d, place_height)
+
+
+| shape_type: str
+| (required) the 4 options available are:
+| "GEOM_SPHERE", "GEOM_BOX", "GEOM_CYLINDER", "GEOM_CAPSULE".
+
+| dim: numpy array of list
+| (optional), if dimensions is specified the length dependents on the shape_type:
+|     GEOM_SPHERE,    dim=[radius],
+|     GEOM_BOX,       dim=[width, length, height],
+|     GEOM_CYLINDER,  dim=[radius, length],
+|     GEOM_CAPSULE,   dim=[radius, length],
+
+| mass: int or float
+| (optional) mass refers to the objects mass, which in uniformly distributed.
+
+| poses_2d: [np.ndarray, ...] or [list, ...]
+| (optional) if specified poses_2d must be a list containing lists or numpy arrays of length 3
+| correspond to [x_position, y_position, orientation] indicates the shapes center of mass.
+| multiple walls can be placed by having multiple pose_2d in poses_2d.
+
+| place_height: int or float
+| (optional) if specified it refers to the z_position of the center of mass
+| if not specified, the shape will be placed such that the shapes bottom
+| aligns with the ground plane.
+
+URDF files
+---------------
+todo
+
+
+Sensors
+--------
+
+A robot can be given a lidar or obstacle sensor by creating a sensor
+object and passing it to the environment:
+
+.. code:: python
+
+    sensor = ObstacleSensor()
+    env.add_sensor(sensor)
+
+The observations from the sensor are returned by the ``env.step(action)`` call.
+The structure of the observation varies depending on the sensor and its arguments
+for more info see the Sensor class and subclasses located at urdfenvs/sensors/
+
+
+Goals
+------
+todo..

docs/source/dependencies.rst

@@ -0,0 +1,30 @@
+Dependencies
+============
+
+Scenes
+-------
+todo...
+
+
+Robot control with the keyboard
+-------------------------------
+
+Control robot actuators with keyboard keys. This is done by:
+
+- Setting up a parent en child process with a pipe connection inbetween
+- Setup and start main process with parent\_connection as argument
+- Setup Responder object with child\_connection as argument
+- Start Responder with parent process as argument
+
+In the main loop an request for action should be made followed by
+waiting for a response as such:
+
+.. code:: python
+
+    parent_conn.send({"request_action": True})
+    keyboard_data = parent_conn.recv()
+    action = keyboard_data["action"]
+
+Additionally custom key bindings and a default action can and passed as arguement
+to the responder. An example can be found in `urdfenvs/examples/keyboard_input.py
+<https://github.com/maxspahn/gym_envs_urdf/blob/master/examples/keyboard_input.py>`_.

docs/source/getting_started.rst

@@ -1,10 +1,34 @@
-Getting started
+Generic URDF robots
 ===================
 
+In this package, generic urdf robots and a panda gym environment are
+available. The goal is to make this environment as easy as possible to
+deploy. Although, we used the OpenAI-Gym framing, these environments are
+not necessarly restricted to Reinforcement-Learning but rather to local
+motion planning in general.
+
+.. |img1| image:: img/pointRobot.gif
+.. |img2| image:: img/pointRobotKeyboardInput.gif
+.. |img3| image:: img/boxerRobot.gif
+.. |img4| image:: img/tiagoKeyboardInput.gif
+.. |img5| image:: img/panda.gif
+.. |img6| image:: img/albert.gif
+
+
++--------+--------+--------+
+| |img1| | |img2| | |img3| |
++--------+--------+--------+
++--------+--------+--------+
+| |img4| | |img5| | |img6| |
++--------+--------+--------+
+
+Getting started
+===============
+
 This is the guide to quickle get going with urdf gym environments.
 
 Pre-requisites
-----------------
+--------------
 
 - Python >3.6, <3.10
 - pip3
@@ -14,7 +38,7 @@ Pre-requisites
 Installation
 ------------
 
-You first have to downlad the repository
+You first have to download the repository
 
 .. code:: bash
 
@@ -46,53 +70,37 @@ The virtual environment is entered by
 
 Inside the virtual environment you can access all the examples.
 
-Examples
------------
-
-Run example
-^^^^^^^^^^^
+Installing dependencies
+-----------------------
 
-You find several python scripts in `examples/
-<https://github.com/maxspahn/gym_envs_urdf/tree/master/examples>`_. You can
-test those examples using the following (if you use poetry, make sure to enter the virtual
-environment first with ``poetry shell``)
+This package depends on casadi for dynamics generation and gym.
+Dependencies should be installed through pip or poetry, see below.
 
-.. code:: python
+Using pip, you can use
 
-   python3 pointRobot.py
-
-Replace pointRobot.py with the name of the script you want to run.
+.. code:: bash
 
-Use environments
-^^^^^^^^^^^^^^^^
+    pip3 install '.[options]'
 
+Using poetry
 
-In the ``examples``, you will find individual examples for all implemented 
-robots. Environments can be created using the normal gym syntax. 
-Gym environments rely mostly on three functions
+.. code:: bash
 
-- ``gym.make(...)`` to create the environment, 
-- ``gym.reset(...)`` to reset the environment, 
-- ``gym.step(action)`` to step one time step in the environment.
+    poetry install -E <options>
 
-For example, in `examples/pointRobot.py
-<https://github.com/maxspahn/gym_envs_urdf/blob/master/examples/pointRobot.py>`_, you 
-can find the following syntax to ``make``, ``reset`` and ``step`` the environment.
+Options are ``keyboard`` and ``scenes``.
 
-.. code:: python
 
-    env = gym.make('pointRobotUrdf-vel-v0', dt=0.05, render=True)
-    ob = env.reset(pos=pos0, vel=vel0)
-    ob, reward, done, info = env.step(action)
+Examples
+-----------
 
-The id-tag in the ``make`` command specifies the robot and the control type.
-You can get a full list of all available environments using
+You find several python scripts in `examples/
+<https://github.com/maxspahn/gym_envs_urdf/tree/master/examples>`_. You can
+test those examples using the following (if you use poetry, make sure to enter the virtual
+environment first with ``poetry shell``)
 
 .. code:: python
 
-   from gym import envs
-   print(envs.registry.all())
-
-
-Go ahead and explore all the examples you can finde there.
+   python3 pointRobot.py
 
+Replace pointRobot.py with the name of the script you want to run.

docs/source/index.rst

@@ -12,8 +12,10 @@ Welcome to urdfEnvs's documentation!
 
    getting_started
    windows
-   introduction
-   developers
+   introduction_to_envs
+   adding_to_envs
+   developpers
+   dependencies
    contributing
    api
 

docs/source/introduction_to_envs.rst

@@ -0,0 +1,54 @@
+Creating environments
+^^^^^^^^^^^^^^^^^^^^^
+
+
+Examples can be found in the `urdfenvs/examples<https://github.com/maxspahn/gym_envs_urdf/tree/master/examples>`_ folder
+you will find individual examples for all implemented
+robots. Environments can be created using the normal gym syntax.
+Gym environments rely mostly on three functions
+
+- ``gym.make(...)`` to create the environment,
+- ``gym.reset(...)`` to reset the environment,
+- ``gym.step(action)`` to step one time step in the environment.
+
+For example, in `examples/pointRobot.py
+<https://github.com/maxspahn/gym_envs_urdf/blob/master/examples/pointRobot.py>`_, you
+can find the following syntax to ``make``, ``reset`` and ``step`` the environment.
+
+.. code:: python
+
+    env = gym.make('pointRobotUrdf-vel-v0', dt=0.05, render=True)
+    ob = env.reset(pos=pos0, vel=vel0)
+    ob, reward, done, info = env.step(action)
+
+The id-tag in the ``make`` command specifies the robot and the control type.
+You can get a full list of all available environments using
+
+.. code:: python
+
+   from gym import envs
+   print(envs.registry.all())
+
+Go ahead and explore all the examples you can find there.
+
+Switching
+--------
+
+Environments can be created using the normal gym syntax. For example the
+below code line creates a toy robot with 3 links.
+Actions are velocities to the individual joints.
+
+.. code:: python
+
+    env = gym.make('nLink-urdf-reacher-vel-v0', n=3, dt=0.01, render=True)
+
+A holonomic and a differential drive mobile manipulator are implemented:
+
+.. code:: python
+
+    env = gym.make('albert-reacher-vel-v0', dt=0.01, render=True)
+    env = gym.make('mobile-reacher-tor-v0', dt=0.01, render=True)
+
+For most robots, different control interfaces are available, velocity
+control, acceleration control and torque control.
+

examples/boxer.py

@@ -13,7 +13,7 @@ def main():
     for e in range(n_episodes):
         ob = env.reset(pos=pos0)
         print(f"Initial observation : {ob}")
-        env.set_walls(limits=[[-4, -4], [4, 4]])
+        env.add_walls()
         print("Starting episode")
         for i in range(n_steps):
             action = defaultAction

examples/point_robot.py

@@ -17,7 +17,7 @@ def main():
     vel0 = np.array([1.0, 0.0, 0.0])
     ob = env.reset(pos=pos0, vel=vel0)
     print(f"Initial observation : {ob}")
-    env.set_walls(limits=[[-3, -2], [3, 2]])
+    env.add_walls()
     if obstacles:
         from examples.scene_objects.obstacles import (
             sphereObst1,