Abstract: 

The rapid convergence of artificial intelligence (AI) and Synthetic Aperture Radar (SAR) is revolutionizing how we perceive, interpret, and generate radar data, unlocking unprecedented capabilities for complex environments. This session invites researchers to explore the transformative potential of AI in redefining SAR’s core challenges: ​high-fidelity imaging, ​robust target recognition, and ​scalable data generation. By integrating physics-driven SAR principles with cutting-edge machine learning—from deep unrolling networks and self-supervised paradigms to generative models—this session aims to bridge the gap between theoretical innovation and real-world deployment. Contributions will highlight how AI augments SAR systems to overcome limitations in sparse data adaptation, cross-domain generalization, and computational efficiency, while fostering interpretability and autonomy. Topics span next-generation imaging architectures, label-efficient recognition frameworks, and synthetic-to-real data generation strategies, with applications in environmental monitoring, defense, and disaster response. We welcome submissions that push the boundaries of AI-SAR synergy, whether through novel algorithms, cross-disciplinary methodologies, or benchmark datasets. Join us in shaping the future of intelligent radar sensing, where data-driven intelligence meets electromagnetic physics to solve tomorrow’s challenges today.

Session Chairs: 

Assoc. Prof. Zhongling Huang (Northwestern Polytechnical University, China),
Assoc. Prof. Jingyuan Xia (National University of Defense and Technology, China), 
Prof. Mihai Datcu (University of Politehnica Bucharest, Germany)