This research project explores the application of causal learning methodologies to enhance trust and explainability in tactical autonomous systems. Interns will work on developing and implementing causal discovery and causal inference techniques specifically designed for unmanned aerial systems (UAS) and other autonomous platforms operating in complex tactical environments. The project addresses a critical challenge in military AI systems: understanding not just correlations in sensor data and mission outcomes, but the underlying causal mechanisms that drive autonomous decision-making. By uncovering causal relationships through discovery algorithms and enabling counterfactual reasoning through causal inference, this work aims to make autonomous systems more interpretable to human operators and more robust when encountering scenarios that differ from their training distributions. 

Interns will gain hands-on experience with cutting-edge causal AI techniques while working on problems with direct real-world impact for defense applications. The project involves developing frameworks that integrate multi-modal data sources (imagery, sensor readings, environmental context) with causal graphical models to support explainable AI (XAI) for tactical autonomy. Participants will learn to implement causal discovery algorithms to identify cause-effect relationships in operational data, apply causal inference methods to generate counterfactual explanations for autonomous decisions, and evaluate how these approaches improve human-machine teaming and situational awareness. This interdisciplinary work bridges machine learning, causal reasoning, and military operations research, providing interns with valuable experience in responsible AI development for high-stakes applications where trust and transparency are paramount.