How to extend

There are two main ways to extend the functionality and environments of navground.learning.

Navground Components

On the navground side, we can add components such as

• new behaviors (to imitate)

• new sensors to include in observations

• new scenarios

In particular, each combination of sensor and scenario define a different environment to train policies.

The navground documentation provides guides on how to add components.

Configurations

On the Gymnasium/PettingZoo side, we can subclass

• ActionConfig to translate actions to commands

• ObservationConfig to extract observations from sensing readings and behavior internal states

• Reward to add specific reward functions

Each of these component can be registered by name so to save/load environments to/from YAML.