Handling Concurrency in Behavior Trees

This paper addresses the concurrency issues affecting Behavior Trees (BTs), a popular tool to model behaviors of autonomous agents in video game and robotics industry. BT designers can easily build complex composing simpler ones, which represents key advantage BTs. The parallel composition BTs expresses way combine concurrent that has high potential, since pre-existing results easier than classical control architectures, as finite state machines or teleo-reactive programs. However, rarely use such due underlying problems similar ones faced programming. As result, composition, despite its finds application only simple where designer guarantee absence conflicts by design. In this paper, we define two new nodes tackle show how exploit them create predictable behaviors. addition, introduce measures assess execution performance different design choices affect them. We validate our approach both simulations real world. Simulated experiments provide statistically significant data, whereas real-world applicability method on robots. provided an open-source implementation novel formulation published all source code reproduce numerical examples experiments.