Polarith AI
1.8
There are several different types of components.
First, there are the system components which are responsible for connecting everything and to do the complex stuff behind the scenes.
Second, we have the Planar Sensor and Spatial Sensor assets which can be associated with a Context component to configure how an agent can observe its world.
For automatic optimization, the Lod Group component can be used.
Third, the components related to the perception pipeline determines what parts of the virtual world can be seen by agents. These should be utilized for organizing your scenes and for centralizing the perception of all agents in order to increase the performance a lot. Due to the fact that behaviours provide their own possibilities for receiving perceptional data, these components are completely optional but highly recommended to be used.
Fourth, the inbuilt path components easily allow the integration of Unity's navigation technology with our steering behaviours. Moreover, Polarith AI provides components for constructing custom path structures with just a few clicks. This way, you can master complex scenarios like patrols and racing tracks very comfortably.
Fifth, there are AI behaviours which determine how to sample the world and to control how agents behave. They are processed by an associated Context component in a specific order and can be divided into the following different subtypes. The concrete order of a behaviour can be set by you. This code is most important to you because it can tell you how you can write behaviours. Thus, we decided to make behaviours open-source on GitHub one after another over time.
Sixth, the formation behaviours that allow you to align your agents in ordered formations. They are special behaviour components that create a magnitude toward their specified position inside the formation. There are different types of formations. You can find the complete source code on GitHub.
At last but not least, the inbuilt character controllers provide the minimalistic possibility to directly see and debug the movement output of the AI system. We used these components to build the example scenes and to show you how the movement decisions made by Polarith AI can be applied.