Secure Perception and Cooperative Control of Autonomous Swarms in Adversarial Scenarios（Yangwen）
Focusing on secure autonomous perception, reliable cooperative control, and trustworthy intelligent decision-making in adversarial swarm scenarios for future autonomous cooperative operations, this research has yielded several pioneering contributions:
l Formulated an unsupervised learning algorithm driven by estimation performance feedback, resolving the critical challenge of attack detection under unknown attack models and scarce data samples.
l Designed swarm navigation and control methodologies for communication-constrained dynamic environments, achieving resilient cooperative control under limited information.
l Proposed a linearly convergent, differentially private gradient tracking algorithm over time-varying topologies, underpinning trustworthy decision-making for large-scale swarms.
The body of work in this area encompasses 62 papers featured in flagship journals such as Automatica and IEEE Transactions. Furthermore, these contributions have garnered one first-prize and two second-prize provincial and ministerial-level awards, providing crucial support for the selection of three individuals into national-level talent programs.
