Intelligence Editor¶

Use Decision Making Preview¶

After making changes to considerations or decisions, you can use our Decision Making Preview feature to check whether the results are as you expected.

For better results, if your inputs have targets, you should create a separate input for each kind of target so that you can change the input value for each target type to see how it affects the decision scores. For example, you can create DistanceToEnemyA, DistanceToEnemyB, DistanceToHealthStation, DistanceToEnergyStation, etc., adjust their values, and then check the result in the Intelligence Editor.

Lock the Intelligence Editor¶

If you want to modify variables from other GameObjects and see how they affect the decision scores, you can lock the Intelligence Editor, select other GameObjects, change those variables in the Inspector Window, and see the results in the Intelligence Editor.

For example, suppose you have an input called DistanceToTarget, and you want to see how it affects the decision scores at runtime. You can lock the Intelligence Editor, drag the target around the current scene, and then check the result in the Intelligence Editor.

For more information, see Lock the Editor.

Group your components into categories¶

As your AI system becomes more complex, you will have many inputs, input normalizations, considerations, and decisions, making it challenging to manage. You should group them into categories for easier management by using CategoryAttribute and CategoryField.

For more information, see Categories.

Organize Fields in the IntelligenceEditor¶

After your defined types become more complex and have a lot of fields, you can organize the fields in the Intelligence Editor by using the Field Attributes.

Update type names and field names in JSON¶

If you change the type or field names of a serializable component, such as inputs, input normalizations, target filters or action tasks; you must also use the JSON Attributes to update those names in the stored JSON.

If you haven’t already, please consider leaving a review on the Asset Store. Whether good or bad, your feedback helps shape the future of this framework, and lets others determine whether it’s a good fit for their games. Thank you so much!💘 I love you all!🥰