Attach_7_-_Task_Order_2_-_Collaborative_Agent-Aided_Teaming_for_Multiple_Operators_of__Multiple_Unmanned_Vehicles_(CAT-MOMU)_SOO.pdf

Task Order 2 - Statement of Objectives Collaborative Agent-Aided Teaming

For Multiple Operators of Multiple Unmanned Vehicles (CAT-MOMU)

1.0 Background

The importance of autonomy for realizing USAF employment of multiple manned and unmanned teamed sensor platforms in future warfighting is well recognized. These new mixed-initiative interactive systems must enable human-machine collaboration and teaming that pairs a human’s pattern recognition and judgement capability with recent machine advances in artificial intelligence and autonomy to facilitate synchronized tactical operations. Agility in tactical decision-making, mission management, and control is also a requirement whereby heterogeneous manned and unmanned systems, with respective machine technologies, jointly problem solve and coordinate task sharing to better meet anticipated complex, ambiguous, and time-challenging warfare conditions.

The USAF has made significant advances in developing intelligent agents and ground station display and control interfaces that will enable a human-agent team to manage multiple unmanned vehicles (UVs) performing a base defense support mission. The approach utilizes a play-based delegation teaming architecture and an intelligent agent-supported Task Manager Interface that assists with needed actions based on ongoing communications. Instantiated in a high-fidelity simulation referred to as IMPACT (Intelligent Multi-UxV Planning with Adaptive Collaborative Control Technologies), this research capability to date has been used to examine the utility of a wide spectrum of control for the management of 12 (simulated) heterogeneous UVs by a single human operator teamed with an intelligent agent. However, to meet envisioned applications involving heterogeneous UVs managed by multiple human operators, the station’s interfaces, as well as the supporting intelligent agent technology and simulation framework need to be expanded to support multi-operator/multi-UV (MOMU) collaboration and provide multi-domain situational awareness/command and control. Eventually, the interfaces and supporting simulation need to be extended to support complex missions that also involve collaboration with manned assets.

2.0 Scope

The present effort will focus on developing and testing interface prototypes that enable at least two human operators, each with initially assigned heterogeneous UVs, to team together to optimally perform a base defense mission. The station’s respective controls and displays, as well as supportive intelligent agents and simulation framework, need to be refined to support evaluation of alternative flexible teaming constructs to support MOMU/Agent collaboration. [Note that determining autonomous triggers for dividing tasks is outside the scope of this effort (e.g., automatically assigned based on ongoing assessment of mission conditions, workload of the operators, etc.).] This effort will address missions where multiple operators (with intelligent agent support) actively coordinate and collaborate in the management of multiple, multi-domain (simulated) UVs. At a minimum, the approach should provide collaboration interfaces for two UV operators. For more complex scenarios, additional operators are acceptable: either more UV operators or other battlefield assets (e.g., sensor operator(s)).

For this task, the existing base defense mission scenario currently supported by the IMPACT simulation will need to be expanded, as well as corresponding types of tasks aligned with the chosen scenario(s). Existing controls, displays, and intelligent agent decision aids will also need to be refined to better support MOMU/Agent collaboration for a multi-domain application. As part of this effort, conduct at least two multi-task, human-in-the-loop evaluations that compare collaborative teaming solutions and/or compare a prototype solution to a representative baseline. It is anticipated that part-task, lower-fidelity simulation/usability evaluations will also be conducted to evaluate alternative approaches to candidate features/designs of interfaces and decision aids.

The results of the repeated cycles of design and evaluation in this task will feed technical efforts that address MOMU for future envisioned complex mission scenarios.

Conduct of this task is anticipated to involve addressing at least the following objectives:

1) Identify the goals/mission tasks/capabilities to be achieved via MOMU/Agent Collaborative Teaming taking into account that future envisioned missions will also involve collaboration with manned assets.

2) Determine how tasking and assets should be shared between two or more multi-UV operators with collaborative teaming.

3) Determine what intelligent agent capabilities/support/decision aids are needed for MOMU/Agent Collaborative Teaming.

4) Determine how operator intent/preference can be accommodated in the teaming solution.

5) Determine what information should be presented in the stations of both operators to support MOMU/Agent Collaborative Teaming.

6) Determine what control functionality is needed in the stations of operators to support MOMU/Agent Collaborative Teaming.

7) Identify enhancements in the simulation’s supporting architecture that are needed for MOMU/Agent Collaborative Teaming in order to: 1) support shared awareness of each operator’s active tasks/plays and resource status, 2) support changes to the existing base defense scenario to include scenario(s)/tasking(s) that require more collaboration, and 3) provide needed simulation control and data recording functionality via the test operator’s console.

8) Identify issues relevant to evaluating different approaches for Human/Agent Collaborative Teaming.

The results of addressing each of the above objectives will identify candidate solutions as well as identify required hardware/software changes to the IMPACT stations and test operator’s console (including intelligent agents, simulation framework, scenarios/taskings, controls, displays, and decision aids). This effort also involves implementing solutions (guided by addressing the above eight objectives) with the purpose of demonstrating and evaluating advanced MOMU/Agent Collaborative Teaming approaches/capabilities.

3.0 Technical Requirements

• Survey the IMPACT simulation as well as other publications and research simulations relevant to future MOMU/Agent Collaborative Teaming applications.

o Conduct technology surveys to identify candidate operator station technologies (displays, controls, and decision aids) that may be leveraged to enhance operator performance and mission effectiveness, and limit operator workload to manageable levels for the targeted MOMU/Agent Collaborative Teaming.

o Survey summaries of UV and manned control stations projects for information related to collaborative teaming interfaces. Evaluate scenarios and mission taskings that are useful to demonstrate the envisioned collaborative teaming and serve as design and evaluation tools to assess competing alternative solutions.

o Assess current and future enabling technologies that will influence the command and control of UVs, such as datalinks, sensors, weapons, and vehicle flight/mission management systems to identify the appropriate technologies and their performance requirements to inform scenario development.

• Conduct a requirements analysis using relevant mission scenarios to define work, functional, information, and control requirements.

o Characterize the mission(s) by providing a detailed description and identifying objective and subjective measures of MOMU/Agent Collaborative Teaming and mission effectiveness and performance.

o Consider the effects of various controls, displays, and decision aid concepts on information and control requirements.

• Develop interface concepts for MOMU/Agent Asset Collaborative Teaming in completion of mission scenarios/taskings.

o For the human-in-the-loop simulation evaluations, identify and describe: 1) physical characteristics of the operator station, 2) control and display design approach, 3) display format symbology and its mechanization, 4) control functionality and its mechanization, 5) proposed decision aiding, automation and/or advanced technology concepts to be incorporated, and

6) enabling technologies and the emulation requirements.

o Define simulator hardware and software requirements and utilize those requirements to implement and/or refine those components in the current IMPACT simulation and supporting Fusion architecture for the ground stations and test operator’s console to be used in evaluations.

• Develop analysis plans for simulation evaluations o Review prior related research and describe a proposed methodology for evaluating the intuitiveness, operability, usability, and effectiveness of controls, displays, and decision aids in simulated mission contexts.

o Define test objectives, simulation capability requirements, test environment/scenario, experimental design, test participant requirements, test matrix, training procedures, measures to be recorded (e.g., performance, mission effectiveness, workload, situation awareness, trust judgement/action, and operator-agent teaming efficiency), and data reduction and analysis procedures.

o Include materials required for the test approval processes, to include institutional, safety, and technical review boards.

• Develop hardware and software architecture designs that integrate interfaces and network protocols into advanced operator station concepts. Develop systems engineering designs and conceptual hardware configurations utilizing specialized CSIL simulation hardware and equipment to meet test program objectives.

• Conduct simulation evaluations including the collection and analysis of the data.

o Acquire and setup special laboratory and test facility equipment for each evaluation, including implementation and integration of simulation features, intelligent agents, and operator interfaces, scenario tasking/procedures, and data collection capabilities to meet the evaluation’s objectives. Verify their implementation prior to the conduct of evaluations. It is anticipated that the evaluations will be conducted in the RHC Crew Systems Integration Laboratory (CSIL) and will involve the use of software packages, models, and networking that provide optimal research capability.

o Conduct evaluations as formal research using accepted experimental design processes. For some research topics, usability assessments are more appropriate and should be conducted using accepted industry practices.

Example usability assessments include user surveys, design (cognitive) walks-through, think-aloud protocols, paper and pencil evaluations, expert evaluations, usability testing, and field studies.

o Analyze objective and subjective data from the evaluations and summarize qualitatively and quantitatively as appropriate, using accepted industry practices (e.g., graphical procedures, parametric repeated measures Analysis of Variance, and non-parametric statistical methods).

o Document the methodology and results from data analyses for each evaluation. Also prepare documentation describing the implemented simulation features and intelligent agents. Prepare materials in support of presentations and publications.

4.0 Operations Security (OPSEC)

The contractor shall provide OPSEC protection for all sensitive/critical information as defined by AFI 10-701 (Operations Security), the 711 HPW OPSEC Plan, and critical information list. The contractor shall participate in the 711 HPW sustained OPSEC awareness training or include OPSEC training as part of their ongoing security program. The 711 HPW OPSEC coordinator will evaluate the OPSEC posture of AF contract activities and operations.

5.0 Additional Requirements

• Highest Security Classification: Unclassified

• Deliverables o Reports: CDRLs A001, A002, A003, A004, A005, A006, A007, A008, A009, A010, A012, A022 o Software: Laboratory software will be developed and delivered o Hardware: Lab prototype hardware

• ITAR: Applicable

• Available Base Support/Facilities: The contractor will be provided access to the following Building 146 laboratory facility to conduct research:

o Crew Systems Integration Laboratory (CSIL)

• GFP/GFI: None

• Flight Test: No

• Human Use: Yes

• Animal Use: No

• Hazardous Materials: No

• Radioactive Materials: No

• Period of Performance: 45 months technical plus 3 months for Final

Technical Report (48 month POP)

• Estimated Funding Profile: The estimated funding for this task order is:

Fiscal Year FY1 FY2 FY3 FY4 Total

Total Amount $1.5M $1.5M $1.0M $1.0M $5.0M