Question

In a Behavior Tree, how does a &#x27;selector&#x27; node differ from a &#x27;sequence&#x27; node when managing the execution of child tasks?

Accepted Answer

A selector node and a sequence node are both control flow nodes that manage how an agent processes child tasks, but they use different logic to determine success or failure. A sequence node acts like an AND gate, requiring every child task to succeed. It executes its children one by one from left to right; if any child task fails, the sequence immediately stops and reports a failure to its parent. If all children succeed, the sequence node reports success. In contrast, a selector node acts like an OR gate, where the goal is to find at least one successful task. The selector executes its children from left to right and stops as soon as any child succeeds, reporting success to its parent. If a child task fails, the selector simply moves on to the next child in the list. The selector only reports a failure if every single one of its children fails. For example, a robot using a sequence to make coffee would need to boil water, grind beans, and pour water in that order, failing if any step fails. Conversely, a robot using a selector to find an object might check a shelf, then a box, then a table; it succeeds as soon as it finds the object in any of those locations, only failing if the object is not in any of them.

Home → All Courses → Engineering and Technology Courses → Robotics and Autonomous Systems → Flashcard

In a Behavior Tree, how does a 'selector' node differ from a 'sequence' node when managing the execution of child tasks?