Guaranteed Performance Control Method For Hypersonic Flight Vehicle With Limited Information

As a space vehicle with strong penetration and maneuverability,the hypersonic flight vehicles(HFVs)have been widely concerned by the world’s military powers.Although huge development has been made on the research of HFVs in recent years,the design of control systems still faces great challenges due to their inherent dynamics and complex constraints.In this paper,the tracking and cooperative control of the HFVs under the condition of information limitation are studied.The main research contents are listed as follows.(1)The trajectory tracking control problem of HFVs with limited model information is studied.According to different model processing methods,based on the appointed-time performance function and the sliding mode control theory,a robust integral controller and a disturbance compensation controller are proposed to realize the trajectory tracking control of HFVs.Additionally,the extended state observer is introduced to deal with the problem of limited model information.Moreover,the introduction of the appointed-time performance function makes the tracking errors converge to the expected tracking accuracy within the userspecified time,which further improves the transient and steady state performance of the system.(2)The attitude tracking control problem of HFVs with limited model information is studied.Based on the backstepping framework and the appointed-time performance function,two attitude tracking control algorithms with guaranteed performance is proposed under the time/eventtriggered mechanism.Wherein,in both algorithms,an extended state observer is introduced to solve the problem of model information limitation caused by the model uncertainties and external disturbances.For the method based on the time-triggered mechanism,in order to avoid the problem of “explosion of terms” caused by the repeated differentiation of the virtual control input in the traditional backstepping control technology,the virtual control law is not introduced in the design of the algorithm,which reduces the online computation burden greatly.Furthermore,for the method based on the event-triggered mechanism,an event-triggered mechanism is introduced to ensure that the controller update would be carried out only when needed,so as to reduce the action frequency of the actuator and reduce its loss.Additionally,the introduction of the appointedtime performance function can avoid the degradation of system performance caused by aperiodic controller update in the event-triggered mechanism and ensure the appointed-time convergence of the tracking error.(3)The cooperation control problem of HFVs with limited model information is studied.According to different model processing methods,based on the appointed-time performance function and the backstepping framework,two formation control methods are proposed for HFVs.In this way,the formation tracking errors could converge to the required precision within the user-specified time,and the desired formation configuration could be realized and maintained.In addition,an extended state observer is introduced to estimate the unmeasurable states and external disturbances of the HFVs,which solves the problem of limited model information.And the adaptive law is used to estimate and compensate the observation errors online.Furthermore,considering the distance constraints between formation members,a new artificial potential function is proposed,so that no collision occurs during formation.(4)The formation control problem of HFVs with limited model information and communication bandwidth is studied.Based on the eventtriggered mechanism and the backstepping framework,an adaptive formation control method is proposed to achieve the desired formation tracking.Wherein,the extended state observer is used to solve the problem of model information limitation.Also,considering the existence of limited communication bandwidth,a dynamic event-triggered mechanism is proposed,which greatly reduce the number of information transmission between formation members.Meanwhile,the appointed-time performance function is designed,which can avoid the degradation of system performance caused by aperiodic controller update in event-triggered mechanism and makes the formation tracking errors converge to the expected accuracy within the user-specified time.Furthermore,in order to achieve collision-free formation tracking,a novel artificial potential function in exponential form is proposed,and the corresponding collision avoidance control term is derived to guarantee a stable and safe formation process.