CHARLES RIVER ANALYTICS, INC. — Department of Defense SBIR Phase I: N162-126

Description

Advanced supervisory control tools are needed for managing large teams of swarms. To address this need, we will design and demonstrate Advanced Swarm System Interface Software Toolkit (ASSIST) for efficient and effective multi-level awareness and proactive management of swarm systems. ASSISTs rich decision-support interfaces will convey key critical information to enable awareness, projection, and effective control inputs across multiple scales of system resolution and abstraction. First, we will expand prior analyses of swarm operations to characterize operators need for observability and directability of automated planning tools. Second, we will leverage this analysis to design a set of novel interface techniques that enhance the operators situation awareness and projection of swarm system behaviors across multiple perspectives and levels of abstraction. These interfaces will extend common interface methods from real-time-strategy games to address specific swarm system challenges. We will also adapt advanced swarm algorithms to better support the operator in understanding when and how interaction should happen with system entities. Finally, we will extend in-house software tools to rapidly prototype ASSIST swarm supervisory control interfaces. We will use these prototypes to demonstrate and evaluate our approaches and to deeply explore the complexities of supervising swarm systems.