Merge pull request #608 from ARISE-Initiative/fix-doc-links

Fix doc links

docs/modules/controllers.rst

@@ -33,7 +33,7 @@ The "WholeBodyIK" composite controller takes in end effector targets, and conver
 ***********************
 
 Third-party controllers integrate custom or external control algorithms into robosuite. Examples include https://github.com/kevinzakka/mink. We provide 
-an example of adding a third part controller in https://robosuite.ai/docs/tutorials/add_controller.html.
+an example of adding a third-party controller in https://robosuite.ai/docs/tutorials/add_controller.html.
 
 
 Workflow of Loading Configs

docs/tutorials/add_environment.md

@@ -1,8 +1,8 @@
 ## Building Your Own Environments
 
-**robosuite** offers great flexibility in creating your own environments. A [task](modeling/task) typically involves the participation of a [robot](modeling/robot_model) with [grippers](modeling/robot_model.html#gripper-model) as its end-effectors, an [arena](modeling/arena) (workspace), and [objects](modeling/object_model) that the robot interacts with. For a detailed overview of our design architecture, please check out the [Overview](modules/overview) page in Modules. Our Modeling APIs provide methods of composing these modularized elements into a scene, which can be loaded in MuJoCo for simulation. To build your own environments, we recommend you take a look at the [Environment classes](simulation/environment) which have used these APIs to define robotics environments and tasks and the [source code](https://github.com/ARISE-Initiative/robosuite/tree/master/robosuite/environments) of our standardized environments. Below we walk through a step-by-step example of building a new tabletop manipulation environment with our APIs.
+**robosuite** offers great flexibility in creating your own environments. A [task](../modeling/task) typically involves the participation of a [robot](../modeling/robot_model) with [grippers](../modeling/robot_model.html#gripper-model) as its end-effectors, an [arena](../modeling/arena) (workspace), and [objects](../modeling/object_model) that the robot interacts with. For a detailed overview of our design architecture, please check out the [Overview](../modules/overview) page in Modules. Our Modeling APIs provide methods of composing these modularized elements into a scene, which can be loaded in MuJoCo for simulation. To build your own environments, we recommend you take a look at the [Environment classes](../simulation/environment) which have used these APIs to define robotics environments and tasks and the [source code](https://github.com/ARISE-Initiative/robosuite/tree/master/robosuite/environments) of our standardized environments. Below we walk through a step-by-step example of building a new tabletop manipulation environment with our APIs.
 
-**Step 1: Creating the world.** All mujoco object definitions are housed in an xml. We create a [MujocoWorldBase](source/robosuite.models) class to do it.
+**Step 1: Creating the world.** All mujoco object definitions are housed in an xml. We create a [MujocoWorldBase](../source/robosuite.models) class to do it.
 ```python
 from robosuite.models import MujocoWorldBase
 
@@ -28,7 +28,7 @@ mujoco_robot.set_base_xpos([0, 0, 0])
 world.merge(mujoco_robot)
 ```
 
-**Step 3: Creating the table.** We can initialize the [TableArena](source/robosuite.models.arenas) instance that creates a table and the floorplane
+**Step 3: Creating the table.** We can initialize the [TableArena](../source/robosuite.models.arenas) instance that creates a table and the floorplane
 ```python
 from robosuite.models.arenas import TableArena
 
@@ -37,7 +37,7 @@ mujoco_arena.set_origin([0.8, 0, 0])
 world.merge(mujoco_arena)
 ```
 
-**Step 4: Adding the object.** For details of `MujocoObject`, refer to the documentation about [MujocoObject](modeling/object_model), we can create a ball and add it to the world.
+**Step 4: Adding the object.** For details of `MujocoObject`, refer to the documentation about [MujocoObject](../modeling/object_model), we can create a ball and add it to the world.
 ```python
 from robosuite.models.objects import BallObject
 from robosuite.utils.mjcf_utils import new_joint
@@ -50,7 +50,7 @@ sphere.set('pos', '1.0 0 1.0')
 world.worldbody.append(sphere)
 ```
 
-**Step 5: Running Simulation.** Once we have created the object, we can obtain a `mujoco_py` model by running
+**Step 5: Running Simulation.** Once we have created the object, we can obtain a `mujoco.MjModel` model by running
 ```python
 model = world.get_model(mode="mujoco")
 ```