Except for the velocity and attitude tracking control problem discussed in the previous six chapters, the cooperative formation is another core aspect of multi-UAV control and serves as the foundation of executing flight missions. In complex flight environment, UAV formation would be affected by external disturbances, such as gusts and electromagnetic interference, as well as model uncertainty, which significantly influences the safety and stability of cooperative flight among multiple UAVs. In response to this challenge, this chapter designs a learning-based optimal control methodology to ensure the performance robustness and optimality of multi-UAV system in the face of strong disturbances.