Papers
1
Total Citations
5
H-Index
1
About
Yukun Ma’s research lies at the intersection of underwater robotics, computer vision, and adversarial machine learning, with a focus on solving the unique challenges of autonomous underwater vehicles (AUVs). Their most cited work, “Adversarial learning-based method for recognition of bionic and highly contextual underwater targets” (2023, 5 citations), tackles a critical problem: distinguishing bionic underwater robots—which mimic real marine life in appearance and behavior—from actual creatures. This is a high-stakes task for AUVs operating in surveillance, ecological monitoring, or defense contexts. Ma’s contribution is a novel adversarial learning framework that enhances recognition accuracy while addressing the large model sizes that plague conventional underwater detection systems. By integrating contextual cues and adversarial training, their method improves robustness against deceptive, biomimetic targets. This work not only advances autonomous underwater perception but also offers a scalable solution for real-time deployment on resource-constrained AUVs. Ma’s research is particularly notable for bridging the gap between adversarial robustness and practical underwater applications, a niche with growing importance as bionic threats become more sophisticated. Their findings provide a foundation for safer, more reliable autonomous operations in complex marine environments.
Research Focus
Key Achievements
Top Papers
- 1