Design Of Flight Control System And Research Of Adaptive Control Algorithm For Small UAV

Small fixed-wing UAV flight control can be divided into three phases: the autonomous take-off phase, the automatic cruise phase, and the autonomous landing phase. The complex control process of the three stages involves the attitude control, height and speed control and navigation control of a UAV. However, in every flight stage, there are many unknown parameter uncertainties and disturbances, so designing a reliable flight control system is very important. In this paper, we design a reliable autopilot hardware system, and realized the basic PID control method for the UAV. In addition, for longitudinal dynamics of Fixed-wing Unmanned Aerial Vehicle(UAV) with strong disturbances and parameter uncertainties, a new High Performance L1 Adaptive Control(HPL1AC) method is proposed. HPL1 AC is based on a combination of the Pole Assignment Controller(PAC) and the L1 adaptive controller, where PAC is utilized to stabilize the closed-loop system and L1 adaptive controller is adopted to guarantee control performance in the presence of parameter uncertainties and disturbances. It is worth mentioning that the BGF(Bounded-Gain Forgetting) estimator is introduced to realize time-varying adaption for the adaptive gain matrix. Finally, simulation results show that the proposed HPL1 AC algorithm achieves desirable transient performance and robustness for shot period dynamics of UAV. The main research content is as follows:First of all, a reliable autopilot system is designed by using modern and advanced sensors as well as other reliable chips. Through the debugging process and related experimental tests, the small UAV control requirements are met. Secondly, by analyzing the UAV flight principle, we design the autonomous take-off and landing control based on PID, strap-down inertial navigation control, flight control methods and attitude control. Then,(HPL1AC) algorithm is proved to be stable by using a method based on the Lyapunov theory. We validate the performance of the algorithm proposed in this paper is better than PID controller through the design of ball and beam system controller experiment. In addition, simulation results of UAV attitude control and take-off and landing controller show that the proposed HPL1 AC algorithm obtains desirable transient performance and robustness for shot period dynamics of UAV.