DON26BZ01-NV032 TITLE: Automated Assessment and Adaptive Training for Simulated Fire Support Coordination

OUSW (R&E) CRITICAL TECHNOLOGY AREA(S): Applied Artificial Intelligence (AAI)

COMPONENT TECHNOLOGY PRIORITY AREA(S): Human-Machine Interfaces;Trusted AI and Autonomy

PROJECTED CMMC LEVEL REQUIREMENT: Level 2 (Self)

OBJECTIVE: Design and develop training tools that assess individual performance in a scenario-based fire support simulator and adapt instruction/scenarios based on that assessment without the need of an instructor in the loop.

DESCRIPTION: Recent Marine Corps publications have emphasized that effective fires employment remains a critical element of Marine Corps lethality and readiness [Refs 2, 3]. Fire support coordination (FSC) is the complex process of planning, integrating, and synchronizing the delivery of indirect fires (e.g., artillery, mortars) and close air support (CAS) to assist maneuver forces on the battlefield. At the company level, fire support is executed by a Fire Support Team (FiST) composed of several members, such as a FiST lead, Forward Observer (FO), Fire Support Officer (FSO), Joint Forward Observer (JFO), and Joint Terminal Attack Controller (JTAC). Some of these roles have a prescribed training pipeline (e.g., JFO, JTAC), whereas others do not (e.g., FiST Lead). The focus of the training for these roles is on individual skill development rather than team-based, integrated execution of fire support in support of a maneuver element. Opportunities for Marines to train collectively in combined arms integration are limited currently. Simulation-based training is available at a few designated locations, but these events require substantial instructor support to simulate different roles across multiple fires agencies and platforms (e.g., Fire Direction Center, CAS aircraft, Ground Force Commander). Live-fire Integrated Training Exercises (ITX), such as Fire Support Coordination Exercises (FSCEX), are costly (e.g., munitions), time and manpower intensive, occur infrequently, and have safety and external agency constraints not present in virtual training (e.g., FAA, host-nation restrictions). Furthermore, both simulated and live training environments require instructors to observe and assess performance with no automated assessment tooling. Across live and simulated events, these assessments are often not standardized and are subjective in nature, limiting opportunities for systematic assessment at the team or individual level. The lack of systematic assessment also limits the ability for Marine Corps to diagnose and address performance impacting the lethality of FiSTs.

Marines need a capability to assess foundational skills in their individual roles (e.g., crawl phase) that is embedded within a realistic fires simulator without requiring instructor facilitation or a full complement of FiST members. Automated assessment within the simulator allows for more objective-based metrics of performance and diagnosis of strengths and weaknesses of individual trainees. That enables schoolhouses and units to track performance on standardized metrics, which could be helpful for readiness assessments. Furthermore, Training and Education 2030 [Ref 3] outlines a student-centered adaptive training solution that tailors the training to the individual Marine based on an assessment of performance and prior research has demonstrated improved student learning outcomes and decreasing time through adaptive training methods – e.g. remediation or adaptive scenario difficulty [Refs 1, 4].

The desired capability of the fire support training solution is to provide tailored training to the individual FiST trainee based on the system’s assessment of performance, targeting skill areas where the trainee is weakest. Providing targeted reps and sets will maximize training time in the crawl phase in addition to improving preparation for the team-based virtual (e.g., walk phase) and live exercises (e.g., run phase). Providing a capability that allows FiST members to practice individual skills ahead of time ensures that trainees can focus on team skill development such as communication and coordination skills with other entities during the time and resource constrained team-based events. Simulation solutions must communicate via standard federated simulation protocols (e.g., DIS6/7, HLA RPR FOM [Refs 5,6]). Preference is given, but not required for submission, to proposals that incorporate or interoperate with existing and/or approved DOW simulation platforms with existing Authority To Operate (ATO) documentation for USMC.

PHASE I: Define and develop a concept for a scenario-based fire support simulation for individual FiST members that incorporates embedded assessment and adaptive training capabilities without the need for an instructor present. The focus of Phase I is developing and demonstrating an individual training solution for a FiST lead while also developing the broader concept for the remaining FiST individual roles. The concept shall include: (1) specific plans for how fundamental skills will be assessed at an individual FiST member level within the simulation and how the simulation will adapt based on the embedded performance assessment (the FiST lead focus / use case must be used to demonstrate a prototype to showcase the maturity and innovation for the proposed solution); (2) an evaluation plan to determine the validity of the assessments and training effectiveness.

PHASE II: Develop a prototype for a scenario-based fire support simulation that incorporates embedded assessment and adaptive training capabilities without the need for an instructor or role players present for the FiST lead role. Develop a plan to expand the assessment and adaptive training capabilities for additional roles in the FiST. Conduct validation of assessments with appropriate end users with coordination assistance from ONR.

Produce the following deliverables: (1) a working prototype of the system with performance data output that can be readable by instructors; (2) supporting software documentation; (3) an assessment validation report that demonstrates the suitability and utility of the training capability in the USMC training pipeline; (4) plans for expanding the assessment and training capabilities to new roles within the FiST.

PHASE III DUAL USE APPLICATIONS: Support the Marine Corps in transitioning the technology for Marine Corps use. Develop the software for evaluation to determine its effectiveness in either a formal Marine Corps schoolhouse or other training setting. As appropriate, focus on broadening capabilities of the training prototype to support additional roles, mission types, and fire support capabilities (e.g., loitering munitions, UAS, etc.).

Applications for dual-use include simulation-based individual and team assessments for training, such as law enforcement and air traffic controllers. The requirement to train and assess large numbers of aircraft controllers to meet personnel shortages has made global headlines. For air traffic control, having objective assessments and simulation-based training is paramount for safety considerations. Other relevant civilian applications that require significant simulation-based resources to train and assess individual and team skills in simulation-based environments include esports, which is a multi-billion-dollar industry with expected continued growth (https://www.statista.com/outlook/amo/esports/worldwide).

REFERENCES:

1. Fraulini, N. W.; Marraffino, M. D.; Garibaldi, A. E.; Johnson, C. I. and Whitmer, D. E. "Adaptive training instructional interventions: A meta-analysis." Military Psychology, July 31, 2024, pp. 1-15. https://pubmed.ncbi.nlm.nih.gov/39083372/
2. United States Marine Corps. (2020). "Force Design 2030." https://www.hqmc.marines.mil/Portals/142/Docs/CMC38%20Force%20Design%202030%20Report%20Phase%20I%20and%20II.pdf
3. United States Marine Corps. (2023, January). "Training and Education 2030. https://www.marines.mil/Portals/1/Docs/Training%20and%20Education%202030.pdf
4. Wickens, C. D.; Hutchins, S.; Carolan, T. and Cumming, J. "Effectiveness of part-task training and increasing-difficulty training strategies: a meta-analysis approach." Human Factors, 55(2), 2013, pp. 461-470. https://pubmed.ncbi.nlm.nih.gov/23691838/
5. "IEEE Standard for Distributed Interactive Simulation--Application Protocols," in IEEE Std 1278.1-2012 (Revision of IEEE Std 1278.1-1995) , vol., no., pp.1-747, 19 Dec. 2012, doi: 10.1109/IEEESTD.2012.6387564
6. SISO-STD-001.1-2015: Standard for Real-time Platform Reference Federation Object Model (RPR FOM), Version 2.0 (10 Aug 2015)

KEYWORDS: Assessment; embedded assessment; adaptive training; fire support; simulation-based training; scenario-based training

---