In the strive for always more safe and autonomous operations for civil aviation the landing phase remains by far the most challenging. Robustifying the already existing automatic landing system, classically based on external infrastructure, with vision-based solutions have recently demonstrated promising perspectives. Nonetheless, the design and validation of such novel algorithms comes at enormous costs and logistic effort for installation, in-flight data collection and testing. This paper introduces a novel Pose-Based Visual Servoing (PBVS) approach for autonomous landing. The proposed method separates the guidance laws from the image flow processing to leverage already certified autopilots. Through computer vision algorithms, a vision-aided Extended Kalman Filter (EKF) receives visual measurements, based on the knowledge of geographical coordinates of some objects of interest, enabling the estimation of the aircraft’s pose with respect to a designated runway. The control framework is implemented for a detailed Boeing 747 model on a real-time simulation platform, which accurately replicates various weather and visual conditions worldwide. The feasibility and performance of the control strategy are evaluated through numerous landings at Toulouse airport, encompassing a range of initial and flight conditions.