All Domain Convergence Applied Research

The All Domain Convergence Applied Research program (PE 0602181A) is a key Army initiative focused on advancing applied research to enable rapid, scalable, and interoperable technologies across all operational domains. The program's overarching goal is to assess the feasibility of emerging technologies in realistic operational environments, leveraging early experimentation to refine approaches and accelerate capability development. This research supports sensor-to-shooter applications from tactical to strategic levels. It employs a system design methodology that aligns with Army experimentation events and the Department of Defense's Combined Joint All-Domain Command and Control (CJADC2) objectives. The program is executed by the Army Research Laboratory (ARL) and complements related efforts in Next Generation Combat Vehicle Technology, Network C3I Technology, and Network C3I Advanced Technology.

Within this program, the Collaborative Convergence Applied Research (CM7) project is the primary line item, with specific objectives aimed at countering adversary technologies and reducing sensor-to-shooter timelines. The project utilizes advanced artificial intelligence (AI) algorithms, decision agent architectures, and data processing methods to improve the speed and effectiveness of mission command in multi-domain operations. The research is designed to accelerate the transition of emerging technologies into operational dominance, supporting the Army's modernization priorities and aligning with the Under Secretary of Defense for Research and Engineering's focus areas.

One major effort under CM7 is AI-Enabled Decision Support in Distributed Networks. This initiative researches techniques to model complex, multi-platform tactical networks in multi-domain operational environments, with the aim of developing robust training datasets for AI-enabled decision support. The work leverages Army doctrine on data value, consensus, uncertainty, and human-agent teaming to optimize decision support training data production. FY 2025 plans include investigating spatio-temporal graph neural networks for adaptive sampling, developing multi-modal analytics for information synthesis, and exploring human-robot distributed decision making using multi-agent reinforcement learning. The project also focuses on explainability and knowledge insertion into neuro-symbolic AI agents.

Another significant component is Synthetic Data for AI-Enabled Decision Support. This research explores the use of synthetic data to augment Army training datasets, optimizing AI performance for rare multi-domain operations targets and environments. Efforts include developing physics-based models and generative adversarial techniques to create synthetic training data, enabling classification of uncommon targets and cost-effective enterprise-level training. FY 2025 activities involve domain adaptation for synthetic-to-real data shifts, experiments with mixed data for 3D mesh representation, and continual learning paradigms to enhance robustness. The project also investigates defensive measures against adversarial AI attacks by modifying asset attributes.

The Data Characterization for AI-Enabled Decision Support effort focuses on improving data management, characterization, curation, labeling, and classification to ensure repeatable and robust AI performance in complex tactical networks. The goal is to enable seamless data mesh connectivity across Department of Defense sources, facilitating continuous improvement of AI algorithms. FY 2025 plans include developing secure processes for transitioning research to laboratory experimentation on mission-relevant data and enabling government research algorithms to inform mission requirement decision makers.

In FY 2026, the Collaborative Convergence Applied Research project is terminated, with funding realigned to other program elements such as Soldier Applied Research and Network C3I Technology. This reflects the completion of key research efforts and a shift toward supporting experimentation strategies, synchronization, and the development of foundational models and learning agents for command and control. The program's accomplishments contribute to the Army's modernization strategy by advancing AI-enabled decision support capabilities for next-generation combat vehicles, networks, future vertical lift, and long-range precision fires.