Automated Human-AI Interaction and Mission Planning with UAS Teams

Purpose and goal

The target is teams of aviation systems such as UAV teams. The problem to be approached is the development of an integrated multi-agent mission planning system that not only coordinates agents and actions to achieve complex mission goals, but also includes interaction actions and data/knowledge producing actions required to achieve such goals. Consequently, physical actions, data/knowledge producing actions, and interaction actions, become part of the mission planning process. The resultant multi-agent mission plans include all three types of actions in a plan when executed.

Expected effects and result

The goal of this project is to provide a formal framework and software system for supporting knowledge intensive interaction among teams of agents, both human and robotic, in relevant scenarios, as a basis for generating mission plans. The primary focus is to support a spectrum of automated mission planning techniques to achieve complex goals. One of the unique novelties of this project is to deeply integrate data and knowledge acquisition and interaction into the actual mission planning and execution processes.

Planned approach and implementation

This project is highly collaborative with its industrial partner, Saab Aeronautics, both in spirit and concretely in terms of specific work packages to involve Saab Aeronautics directly in the research proposed. The project will be pursued in the context of state-of-art national research arenas, and will use sophisticated experimental UAVs and ground robotic systems from both LiU and Saab for active experimentation. This is a unique and important aspect of the project that provides the ability to field test all research ideas using complex, heterogeneous teams of UAVs.