The Army is evolving its warfighting concepts to militarily compete, penetrate, dis-integrate, and exploit adversaries as part of a Multi-Domain Operations (MDO) Joint Force. To solve some of MDO’s most critical problems, the future force requires the ability to converge capabilities from across multiple domains at speeds and scales beyond human cognitive abilities. Thus, disruptive foundational research is required for data, information, computational and network sciences with cross-cutting research in Artificial Intelligence (AI) and Machine Learning (ML) for MDO. As Figure 5 illustrates, the Army faces rapidly changing, never-before-seen situations, where pre-existing concepts, capabilities and training data will quickly become ineffective in complex multi-domain environments with adversaries employing deception. This is especially challenging at the Tactical Edge where the operating environment is dynamic, distributed, resource-constrained, fast-paced, contested, and often physically and virtually isolated.
CCDC ARL’S Network & Information Sciences Foundational research competency and Computational Sciences Foundational research competency are focused on identifying, creating, and transitioning scientific discoveries and technological innovations underlying four research areas: (1) Computational Modeling of Complex Systems, (2) Data Processing and Data Analytics, (3) Intelligent and Autonomous Systems, and (4) Communications, Networks and Cyber. In addition to significantly improving and supporting the Army’s existing warfighting capabilities, these competencies concentrate on high-risk and high-payoff transformational basic research; and critically-focused, promising applied research that are expected to have ground-breaking impacts on the Army’s warfighting capabilities in MDO. The integrated research focus and vision for MDO are as follow:
Research Focus: Develop fundamental sciences of computation and networking to ensure the intelligent, resilient processing & transfer of battlefield information for rapid human and automated decision making in Multi-Domain Operations.
Vision for MDO: Information dominance for the Army’s multi-domain military operations.
CCDC ARL’s research in Network & Information Sciences and Computational Sciences supports ARL Essential Research Programs (ERP) and Army Modernization Priorities by creating Army-unique and disruptive technologies with AI Cross-Cutting Research, while also preventing technological surprise from near-peer adversaries and enabling disruptive capabilities required for success in MDO environments. The three key ERPs supported by Network & Information Sciences and Computational Sciences are Artificial Intelligence for Maneuver and Mobility (AIMM), Foundational Research for Electronic Warfare in Multi-Domain Operations (FREEDOM), and Quantum - Position, Navigation and Timing (Q-PNT). AIMM seeks to integrate autonomous mobility and context-aware decision making to deliver autonomous maneuver designed for MDO environments. FREEDOM’s objective is to create persistent distributed, disaggregated & adaptive electronic warfare (EW) technologies for multi-domain overmatch through combined cyber and electromagnetic research. Finally, Q-PNT will provide accurate and trusted PNT to the Army’s dismounts under contested environments and ensures overmatch by maximizing its freedom of maneuver while creating areas of denial on the battlefield against near-peer threats.
Focuses on gaining a greater understanding of fundamental theories and emerging technologies that support intelligent networks and information systems that perform acquisition, delivery, analysis, reasoning, decision-making, collaborative communication, and assurance of information and knowledge. Insights and innovations gained through these research efforts will lead to technological developments that make it possible to manage, disseminate and utilize information to enable distributed intelligent systems to operate effectively in MDO. Technologies resulting from these efforts will have a direct impact on future Army capabilities in Command, Control, Communications, Computers, Intelligence, Surveillance and Reconnaissance (C4ISR), networks, intelligent systems, and cyber security. The envisioned capability specialty areas for Network & Information Sciences include:
- Adaptive processing and communication for C4I in the tactical environment
- Intelligent systems that can maneuver in complex environment
- Predictive analytics for situational understanding
- Distributed information delivery and processing for decision making
- Defensive cyber
- Environmental modeling & impact characterization for multi-echelon decision making
And the envisioned technical specialty areas include:
- AI & ML with distributed, dinky, dynamic, disparate, dirty data
- Uncertainty characterization and quantification
- Human information interaction
- Adversarial detection, understanding & resilience
- Analysis & prediction of the physical environmental effects
Network & Information Sciences have a number of AI Cross-Cutting Research (AICCR) elements that address key AI research challenges for MDO. It is recognized that the military domains are frequently distinct from commercial applications because of: rapidly changing situations; limited access to real data to train AI and limited resources with size, weight, and power – time (SWAP-T) constraints, and noisy incomplete, uncertain, and erroneous data inputs during operations; and peer adversaries that employ deceptive techniques to defeat algorithms. Thus, the primary goal of AICCR is to research and develop artificially intelligent agents (heterogeneous & distributed) that rapidly learn, adapt, reason and act in contested, austere and congested environments by addressing the following AI & ML research challenges in Network & Information Sciences:
- Learning in complex environments
- AI & ML with small samples, dirty data, high clutter
- AI & ML with highly heterogeneous data
- Adversarial AI & ML in contested, deceptive environment
- Resource-constrained processing at the point-of-need
- Distributed AI & ML with limited communications
- AI & ML computing with extremely low SWAP-T
- Generalizable & predictable AI (Explainability & programmability for AI & ML)