Using Mutli-Agent Systems to Improve the Level of Autonomy for Operators Controlling Unmanned Vehicles

摘要

This paper describes preliminary work performed to gain an understanding of how to implement a Multi-Agent System (MAS) that can be used to improve the Level of Automation (LOA) used by operators when controlling Unmanned Air Vehicles (UAVs). The ultimate goal is to build an Autonomous Vehicle Management System (AVMS) with intuitive controls that empower the operator to achieve more with less effort. This concept will be supported by a framework comprised of inter-operable MASs acting as teams. Each team has a modular design that support dynamically linkable capabilities. The design provides the operator with an interface to enabled them to select the mode of operation desired based on trust. Technology can be used to provide automation, but innovation enables humans to make it trusted, robust, efficient and effective. Technology and research will effect the ability to adapt, therefore this paper presents the current issues relating to the existing state of knowledge in this domain. This premise relates to the fact that machines can't go beyond what they are programmed to achieve.