Fix errors in the docs

docs/challenge/simulation.rst

@@ -1,2 +1,73 @@
 Simulation
-==========
+==========
+
+The first part of the challenge is to complete the race track in simulation. To that end, we provide a high-fidelity simulator with interfaces that are identical to the real-world setup. We use PyBullet as the underlying simulator to model the physics of the drone and to render the environment.
+
+Difficulty Levels
+-----------------
+
+The challenge is divided into different difficulty levels, each specified by a TOML configuration file. These levels progressively increase in complexity:
+
+.. list-table::
+   :header-rows: 1
+   :widths: 20 10 15 20 15 20
+
+   * - Evaluation Scenario
+     - Constraints
+     - Rand. Inertial Properties
+     - Randomized Obstacles, Gates
+     - Rand. Between Episodes
+     - Notes
+   * - Level 0 (config/level0.toml)
+     - Yes
+     - No
+     - No
+     - No
+     - Perfect knowledge
+   * - Level 1 (config/level1.toml)
+     - Yes
+     - Yes
+     - No
+     - No
+     - Adaptive
+   * - Level 2 (config/level2.toml)
+     - Yes
+     - Yes
+     - Yes
+     - No
+     - Learning, re-planning
+   * - Level 3 (config/level3.toml)
+     - Yes
+     - Yes
+     - Yes
+     - Yes
+     - Robustness
+   * - sim2real
+     - Yes
+     - Real-life hardware
+     - Yes
+     - No
+     - Sim2real transfer
+
+.. note::
+   "Rand. Between Episodes" (governed by argument `reseed_on_reset`) determines whether randomized properties and positions vary or are kept constant (by re-seeding the random number generator on each `env.reset()`) across episodes.
+
+You may use the easier scenarios to develop and debug your controller. However, the final evaluation will be on the hardest scenario (Level 3).
+
+Switching Between Configurations
+--------------------------------
+
+You can choose which configuration to use by changing the `--config` command line option. For example, to run the example controller on the hardest scenario, use the following command:
+
+.. code-block:: bash
+
+   python scripts/sim.py --config config/level3.toml
+
+To use your own controller, you can pass the path to your controller script as the `--controller` argument. For example:
+
+.. code-block:: bash
+
+   python scripts/sim.py --config config/level3.toml --controller my_controller.py
+
+.. note::
+    You can also write your controller file directly into the config files located in the `config` folder. That way, you don't need to specify the controller script when running the simulation, and the controller will be used in the automated challenge evaluation.

lsy_drone_racing/control/__init__.py

@@ -6,11 +6,11 @@
 
 To give you an idea of what you need to do, we also include some example implementations:
 
-- BaseController: The abstract base class defining the interface for all controllers.
-- PPO Controller: An example implementation using a pre-trained Proximal Policy Optimization (PPO)
-  model.
-- Trajectory Controller: A controller that follows a pre-defined trajectory using cubic spline
-  interpolation.
+* :class:`~.BaseController`: The abstract base class defining the interface for all controllers.
+* :class:`PPOController <lsy_drone_racing.control.ppo_controller.PPOController>`: An example
+  implementation using a pre-trained Proximal Policy Optimization (PPO) model.
+* :class:`PPOController <lsy_drone_racing.control.trajectory_controller.TrajectoryController>`: A
+  controller that follows a pre-defined trajectory using cubic spline interpolation.
 """
 
 from lsy_drone_racing.control.controller import BaseController

lsy_drone_racing/control/controller.py

@@ -5,8 +5,9 @@
 from that, you are free to add any additional methods, attributes, or classes to your controller.
 
 As an example, you could load the weights of a neural network in the constructor and use it to
-compute the control commands in the `compute_control` method. You could also use the `step_callback`
-method to update the controller state at runtime.
+compute the control commands in the :meth:`compute_control <.BaseController.compute_control>`
+method. You could also use the :meth:`step_callback <.BaseController.step_callback>` method to
+update the controller state at runtime.
 
 Note:
     You can only define one controller class in a single file. Otherwise we will not be able to

lsy_drone_racing/control/ppo_controller.py

@@ -1,4 +1,13 @@
-"""Example implementation of a controller using a pre-trained PPO model."""
+"""Example implementation of a controller using a pre-trained PPO model.
+
+The controller loads the pre-trained weights of the policy and uses it to compute the next action
+based on the current observation.
+
+.. note::
+    You need to install the 
+    `stable-baselines3 <https://stable-baselines3.readthedocs.io/en/master/>`_ library to use this
+    controller.
+"""
 
 from __future__ import annotations  # Python 3.10 type hints
 

lsy_drone_racing/control/trajectory_controller.py

@@ -1,4 +1,13 @@
-"""Controller that follows a pre-defined trajectory."""
+"""Controller that follows a pre-defined trajectory.
+
+It uses a cubic spline interpolation to generate a smooth trajectory through a series of waypoints.
+At each time step, the controller computes the next desired position by evaluating the spline.
+
+.. note::
+    The waypoints are hard-coded in the controller for demonstration purposes. In practice, you
+    would need to generate the splines adaptively based on the track layout, and recompute the
+    trajectory if you receive updated gate and obstacle poses.
+"""
 
 from __future__ import annotations  # Python 3.10 type hints
 

lsy_drone_racing/envs/drone_racing_deploy_env.py

@@ -2,21 +2,19 @@
 
 This module provides environments for deploying drone racing algorithms on physical hardware,
 mirroring the functionality of the simulation environments in the
-:mod:~lsy_drone_racing.envs.drone_racing module.
+:mod:`~lsy_drone_racing.envs.drone_racing_env` module.
 
 Key components:
 
-- :class:~.DroneRacingDeployEnv: A Gymnasium environment for controlling a real Crazyflie drone in a
-   physical race track, using Vicon motion capture for positioning.
-- :class:~.DroneRacingThrustDeployEnv: A variant of DroneRacingDeployEnv that uses collective thrust
-   and attitude commands for control.
+* :class:`~.DroneRacingDeployEnv`: A Gymnasium environment for controlling a real Crazyflie drone in
+  a physical race track, using Vicon motion capture for positioning.
+* :class:`~.DroneRacingThrustDeployEnv`: A variant of :class:`~.DroneRacingDeployEnv` that uses
+  collective thrust and attitude commands for control.
 
 These environments maintain consistent interfaces with their simulation counterparts
-(:class:~lsy_drone_racing.envs.drone_racing_env.DroneRacingEnv and
-:class:~lsy_drone_racing.envs.drone_racing_thrust_env.DroneRacingThrustEnv), allowing for seamless
-transition from simulation to real-world deployment. They handle the complexities of interfacing
-with physical hardware while providing the same observation and action spaces as the simulation
-environments.
+(:class:`~.DroneRacingEnv` and :class:`~.DroneRacingThrustEnv`), allowing for seamless transition
+from simulation to real-world deployment. They handle the complexities of interfacing with physical
+hardware while providing the same observation and action spaces as the simulation environments.
 
 The module integrates with ROS, Crazyswarm, and Vicon systems to enable real-world drone control and
 tracking in a racing scenario.

lsy_drone_racing/envs/drone_racing_env.py

@@ -4,21 +4,23 @@
 between the drone racing simulation and the user's control algorithms.
 
 It serves as a bridge between the high-level race control and the low-level drone physics
-simulation. The environments defined here (DroneRacingEnv and DroneRacingThrustEnv) expose a common 
-interface for all controller types, allowing for easy integration and testing of different control 
-algorithms, comparison of control strategies, and deployment on our hardware.
+simulation. The environments defined here 
+(:class:`~.DroneRacingEnv` and :class:`~.DroneRacingThrustEnv`) expose a common interface for all
+controller types, allowing for easy integration and testing of different control algorithms,
+comparison of control strategies, and deployment on our hardware.
 
 Key roles in the project:
-1. Abstraction Layer: Provides a standardized Gymnasium interface for interacting with the drone
-   racing simulation, abstracting away the underlying physics engine.
-2. State Management: Handles the tracking of race progress, gate passages, and termination
-   conditions.
-3. Observation Processing: Manages the transformation of raw simulation data into structured 
-   observations suitable for control algorithms.
-4. Action Interpretation: Translates high-level control commands into appropriate inputs for the
-   underlying simulation.
-5. Configuration Interface: Allows for easy customization of race scenarios, environmental
-   conditions, and simulation parameters.
+
+* Abstraction Layer: Provides a standardized Gymnasium interface for interacting with the
+  drone racing simulation, abstracting away the underlying physics engine.
+* State Management: Handles the tracking of race progress, gate passages, and termination
+  conditions.
+* Observation Processing: Manages the transformation of raw simulation data into structured
+  observations suitable for control algorithms.
+* Action Interpretation: Translates high-level control commands into appropriate inputs for the
+  underlying simulation.
+* Configuration Interface: Allows for easy customization of race scenarios, environmental
+  conditions, and simulation parameters.
 """
 
 from __future__ import annotations

lsy_drone_racing/sim/__init__.py

@@ -3,20 +3,21 @@
 This module provides a high-fidelity simulation setup for quadrotor drones, particularly focused on
 drone racing scenarios. It includes:
 
-- A physics-based simulation using PyBullet
-- Detailed drone dynamics and control models
-- Symbolic representations for dynamics, observations, and cost functions
-- Configurable noise and disturbance models
+* A physics-based simulation using PyBullet
+* Detailed drone dynamics and control models
+* Symbolic representations for dynamics, observations, and cost functions
+* Configurable noise and disturbance models
 
 The simulation environment allows for realistic modeling of drone behavior, including aerodynamics,
 motor dynamics, and sensor characteristics. It supports both high-fidelity physics simulations and
 analytical models for predictable dynamics.
 
 Key components of the simulation include:
-- Drone state representation and dynamics
-- An exchangeable physics backend with PyBullet as visualizer
-- Customizable environmental factors and disturbances
-- Symbolic models for advanced control techniques
+
+* Drone state representation and dynamics
+* An exchangeable physics backend with PyBullet as visualizer
+* Customizable environmental factors and disturbances
+* Symbolic models for advanced control techniques
 
 The physics backend utilizes PyBullet by default for rigid body dynamics simulation, providing
 accurate modeling of collisions, constraints, and multi-body interactions. It can also be replaced

lsy_drone_racing/sim/drone.py

@@ -8,11 +8,11 @@
 The Drone class is a core component of the simulation environment and is heavily utilized
 in the sim.py module. It handles various aspects of drone behavior, including:
 
-- Initialization of drone parameters and firmware
-- Management of drone state and control inputs
-- Implementation of controller logic
-- Conversion between different units (e.g., thrust to RPM)
-- Handling of sensor data and low-pass filtering
+* Initialization of drone parameters and firmware
+* Management of drone state and control inputs
+* Implementation of controller logic
+* Conversion between different units (e.g., thrust to RPM)
+* Handling of sensor data and low-pass filtering
 
 This module also includes the DroneParams dataclass, which encapsulates the physical
 and inferred parameters of the Crazyflie 2.1 drone.

lsy_drone_racing/sim/sim.py

@@ -5,11 +5,11 @@
 
 Features:
 
-- PyBullet-based physics simulation
-- Configurable drone parameters and initial conditions
-- Support for a single drone (multi-drone support not yet implemented)
-- Disturbance and randomization options
-- Integration with symbolic models
+* PyBullet-based physics simulation
+* Configurable drone parameters and initial conditions
+* Support for a single drone (multi-drone support not yet implemented)
+* Disturbance and randomization options
+* Integration with symbolic models
 
 The simulation is derived from the gym-pybullet-drones project:
 https://github.com/utiasDSL/gym-pybullet-drones

lsy_drone_racing/sim/symbolic.py

@@ -3,10 +3,11 @@
 This module provides functionality to create symbolic representations of the drone dynamics,
 observations, and cost functions using CasADi. It includes:
 
-- SymbolicModel: A class that encapsulates the symbolic representation of the drone model,
-  including dynamics, observations, and cost functions.
-- symbolic: A function that creates a SymbolicModel instance for a given drone configuration.
-- Helper functions for creating rotation matrices and other mathematical operations.
+* :class:`~.SymbolicModel`: A class that encapsulates the symbolic representation of the drone
+  model, including dynamics, observations, and cost functions.
+* :func:`symbolic <lsy_drone_racing.sim.symbolic.symbolic>`: A function that creates a
+  :class:`~.SymbolicModel` instance for a given drone configuration.
+* Helper functions for creating rotation matrices and other mathematical operations.
 
 The symbolic models created by this module can be used for various control and estimation tasks,
 such as model predictive control (MPC) or state estimation. They provide a convenient way to work

lsy_drone_racing/vicon.py

@@ -3,9 +3,9 @@
 It defines the Vicon class, which handles communication with the Vicon system through ROS messages.
 The Vicon class is responsible for:
 
-- Tracking the drone and other objects (gates, obstacles) in the racing environment.
-- Providing real-time pose (position and orientation) data for tracked objects.
-- Calculating velocities and angular velocities based on pose changes.
+* Tracking the drone and other objects (gates, obstacles) in the racing environment.
+* Providing real-time pose (position and orientation) data for tracked objects.
+* Calculating velocities and angular velocities based on pose changes.
 
 This module is necessary to provide the real-world positioning data for the drone and race track
 elements.