Exploring network enabled concepts for U(C)AV payload driven navigation

Abstract

Mission effectiveness requires a bandwidth of the Observe-Orient-Decide-Act (OODA) loop that is large enough to cope with the dynamics of the particular target environment. If not already, the current 'human' interface between the navigation, sensor and weapons functions is likely to become the bottleneck that limits the bandwidth of the OODA loop. Some scenarios foresee a future where this bottleneck is eliminated by removing the human operator from the OODA loop using UAVs that autonomously adjust their goals and plans based on the sensed state of the environment. The OODA loop comprises several nested loops, each with increasing bandwidth. Rather than removing the human operator from the OODA loop altogether, the concept explored in this paper aims to move the human operator to a lower bandwidth loop. The operator is supported by a system that dynamically performs a translation of target-payload parameters to a navigation/guidance plan with a level of execution authority that is determined by the dynamics of the environment. As such it represents an evolutionary approach from current systems that are automated up to the plan execution level and systems that autonomously adjust their goals. In this paper, the concept is explained and the options and rationale for the selection of the appropriate level of automation are discussed. Also, an overview of the current development activities is presented.

A Framework for Multiple Ground Target Finding and Inspection Using a Multirotor UAS

“…Most previous studies used a control system approach to control the UAS. Another possible approach is a cognitive model-based approach of OODA [30,31]. For example, Karim and Heinze [32] implemented and compared two distinct autonomous UAS controllers.…”