SBIR-STTR Award

Several unmanned air vehicle (UAV) programs are investigating ways of automating UAV control system functions to increase the number of aircraft controlled by a single operator. However, research indicates that implementing high-levels of automation and putting the human operator in a passive monitoring role can leave the operator out of the loop and unable to intervene effectively when required to take control of system tasks. Also, significant workload increases can occur as the human assumes system control tasks. To counter these problems, a human-centered approach to automation has been proposed, in which the human and the machine work together as a system. The goal is to keep the human in the loop by having the human perform meaningful tasks and to keep the workload manageable at all times. Two complimentary approaches have been developed that address the OOTL performance problem: Level of Automation and Adaptive Automation. Tools are needed to investigate the feasibility of implementing adaptive levels of autonomy in a UAV control station context. The goal of this SBIR program is to devise an architecture suitable for a multi-UAV control station simulation test-bed that will allow testing and evaluating different methods for adaptive levels of autonomy.

Benefits:
The product of this effort will provide the Air Force with a research test-bed for studies of the roles of humans and autonomy in controlling multiple UAVs and for generating guidelines for autonomous operations. The results of this program can be applied to other DOD unmanned vehicle efforts. In particular, this effort is synergistic with the DARPA UCAV effort and the Global Hawk program, and it supports the Navy’s goals for autonomous UAV operations for vehicles such as the Fire Scout and BAMS. A generalized version of the architecture may be applicable to commercial applications of automation.

Keywords:
Adaptive Autonomy, Levels of Autonomy, UAV, Automation, Autonomous Control Level, Human-Centered Automation

OR Concepts Applied (ORCA) is currently involved in a Phase I SBIR effort for AFRL/HECI entitled Adaptive Levels of Autonomy (ALOA) for UAV Supervisory Control. In Phase II, ORCA will build on our Phase I successes, which are a direct result of years of experience designing mission planning and LO autorouting software tools. In Phase I, ORCA devised the ALOA architecture and defined multiple levels of autonomy (LOA) for four operator tasks: allocation, route planning, imagery analysis, and weapon control. To demonstrate the architecture and the implementation of the LOA, we designed a prototype multi-UAV control station emulator (MCSE) test bed, which utilizes ORCA-developed mission planning software components. ORCA’s extensive internal IR&D has produced state-of-the-art mission planning tools that allow fully autonomous route planning and allocation. Experience from the J-UCAS effort and other UAV programs gives us first hand knowledge of the tools and decision aids operators need for manual planning. This experience allows us to implement the two autonomy extremes for route planning and allocation: manual and fully autonomous. We successfully implemented intermediate LOA in Phase I using aspects of manual and autonomous planning. In Phase II, we will mature the ALOA architecture and finalize the MCSE test-bed design.

Keywords:
Levels Of Autonomy, Adaptive Autonomy, Uav, Automa