[Objective] Compound attitude control based on aerodynamic, thrust vector, direct force, and other manipulation structures is a hot research topic in modern aeronautics and astronautics. Additionally, improving the tracking ability of the system under high altitude and low aerodynamic efficiency is essential. To simplify the design, direct force is generally regarded as a continuous control variable in the study of compound attitude control for aircraft, which is different from the actual discrete engineering characteristics. In view of the large gap between direct force continuous processing and actual engineering characteristics, this study proposes control system construction and control parameter optimization methods based on the discrete characteristics of direct force. [Methods] First, based on the operational characteristics of each manipulation structure, a pulse modulator, which embodies the discrete characteristics of direct force, is used to construct the framework of the attitude control system. The Simulink simulation model of the compound aircraft attitude control system is constructed using MATLAB software according to the designed control system structure and the mathematical model of each link. Second, given the complex characteristics of nonlinear, strong coupling, and multiconstraints for the compound control system, the Particle Swarm Optimization (PSO) algorithm is adopted to optimize the control law and control assignment parameters based on the algorithm framework of intelligent control and intelligent distribution laws. Thus, an intelligent control system is constructed. Third, the research method for intelligent control problems is explained through digital simulation from three aspects: control algorithm, PSO algorithm parameters, and aircraft system parameters. [Results] The simulation results show that the proposed method greatly improves the control performance of the attitude-tracking problem, including steady-state error, rise time, adjustment time, and overshoot. It is obviously superior to the traditional PID control method in terms of tracking precision, speed, and security. Additionally, it is beneficial to overcome the difficulty of adjusting the control parameters caused by the change in the flight parameters. The proposed system can adapt to real-time changes in the flight environment and state, effectively exert the control efficiency of each executing agency, and improve the adaptive control ability of the system. Finally, based on the above research achievements, the experimental system for intelligent aircraft attitude control is designed and developed. This system is applied to the practical teaching of intelligent control theory and process control systems. It can also be used to strengthen students' understanding of the application and innovation of intelligent control theory in actual engineering problems from the aspects of attitude control principle, control structure design, mathematical model building, intelligent control algorithm design and implementation, and algorithm research. [Conclusions] This experimental system combines the engineering problems of aircraft attitude control with experimental teaching. By guiding students to project-based learning independently, it stimulates their interest in learning, effectively improves their ability to combine theory with practice, helps them to master scientific research methods, and cultivates and promotes students' practical and innovative abilities. In this way, an independent and open experimental research platform for the practical teaching of intelligent control courses and the cultivation of innovative talents is provided. [ABSTRACT FROM AUTHOR]