This paper presents an extremum seeking control (ESC) optimization method to enhance aerodynamic efficiency in UAV formation flight through wake vortex optimization. Since drag reduction is challenging to measure directly, the study proposes a performance function integrating wingtip sensor measurements and the Hallock-Burnham (H-B) wake fitting model to optimize relative positions within the formation. Initially, a high-fidelity wake model is developed using computational fluid dynamics (CFD) to accurately represent the leader UAV’s wake impact. The designed performance function incorporates both induced drag and profile drag from trimmed rolling moments, leveraging real-time wake sensing and velocity fitting. Optimization employs a Newton-Raphson method combined with an extended Kalman filter (EKF) for dynamic estimation of the optimal UAV positions. Simulations with the Skywalker X8 UAV confirm the method effectively converges to the optimal formation position, demonstrating its feasibility.