Research On Adaptive Fault-tolerant Cooperative Control Method For Multi-UAVs And Its Applications

In the past few years,fault-tolerant cooperative control(FTCC)of multiple unmanned aerial vehicles(multi-UAVs)has been a topic in the control and aviation communities,and has numerous potential applications in the field of safe cooperative work.Tracking error variation and time interval of fault compensation are two factors closely related to safe cooperative work.Moreover,unknown aerodynamic parameters and external wind disturbances can also affect the FTCC system performance.However,these above problems are rarely addressed.To make the FTCC research more practical,addressing the above problems in the FTCC scheme is necessary.Recently,the rapid developments of cooperative control methods for multiple agents and intelligent control methods have provided new ideas for the research of FTCC schemes for multi-UAVs.However,the integration of cooperative control and intelligent control methods into the development of FTCC schemes is very difficult and very few results have been obtained.To further develop effective FTCC schemes for multiple fixed-wing UAVs,the neural networks(NNs),adaptive mechanism,prescribed performance control(PPC),finite-time control,fractional-order control(FOC)and disturbance observer(DO)methods are utilized in this dissertation to ensure the formation flight safety of multi-UAVs against actuator faults,unknown aerodynamic parameters and wind effects.Compared to other existing works,the main research works and contributions are as follows:(1)To attenuate the adverse effects and restrict the tracking error variations induced by the actuator faults,a PPC-based adaptive FTCC scheme is developed for the longitudinal motions of multi-UAVs.By developing a distributed sliding-mode estimator(SME),the leader UAV’s altitude is estimated for each follower UAV and the stability of the proposed distributed SME is analyzed.Then,based on the estimated altitude references for all follower UAVs,prescribed performance functions are used to transform the altitude tracking errors into a new set of errors.Furthermore,regarding the transformed error dynamics,a PPC-based adaptive FTCC scheme is investigated for multi-UAVs based on the NN and adaptive methods.Finally,simulation results have shown the effectiveness of the proposed control scheme.(2)By using the finite-time theory to compress the time interval of fault compensation,a fault-tolerant containment control scheme is proposed for the longitudinal motions of multiUAVs including multiple leader UAVs and multiple follower UAVs.Based on the basic graph theory,a new distributed finite-time SME is developed for each follower UAV to obtain the weighted averages of leader UAVs’ positions and velocities in finite time.Furthermore,by using the error transformation,NN,adaptive mechanism,and minimum parameter learning of NN(MPLNN),a finite-time adaptive FTCC scheme is developed for the safe containment formation of multi-UAVs against actuator faults and obstacles.Finally,comparative simulation results have shown that the proposed method can achieve higher control accuracy,robustness,and faulttolerant capabilities than the infinite-time control method or the traditional dynamic surface control(TDSC)method.(3)To achieve a safe attitude synchronization tracking control of multi-UAVs against actuator faults,a PPC-based finite-time adaptive FTCC scheme is developed to restrict the tracking errors and compress the time interval of fault compensation.Based on the individual tracking error and synchronization error of each UAV,the attitude synchronization tracking error is first established for each UAV.Then,by using the prescribed performance functions,the attitude synchronization tracking errors of all UAVs are transformed into a new set of errors.Furthermore,based on the transformed errors,a PPC-based finite-time adaptive FTCC scheme is proposed for the attitude synchronization tracking of multi-UAVs by using the finite-time theory,NN,adaptive mechanism,robust differentiator,and MPLNN.Finally,comparative simulation results have revealed the superiority of the proposed control scheme.(4)To counteract the in-flight actuator faults and wind effects during the formation flight,an adaptive fractional-order FTCC(FOFTCC)scheme is investigated for the attitude synchronization tracking control of multi-UAVs.By considering the wind effects and actuator faults,the UAV attitude model is first established and the control-oriented attitude model is also formulated to facilitate the controller design.Then,based on the synchronized tracking error dynamics,an adaptive FOFTCC scheme is proposed for multi-UAVs against actuator faults and wind effects.Moreover,prediction errors,which can be used to reflect the approximation abilities of NN and DO,are also integrated into the adaptive laws and DOs to improve the approximation performance.Finally,simulation results have shown that the proposed control scheme can effectively attenuate the adverse effects induced by the actuator faults and can also provide an extra degree-of-freedom(DOF)to the parameter adjustment.(5)Based on the fuzzy neural networks(FNNs)and FOC theory,an FNN-based adaptive FOFTCC scheme is developed for the attitude tracking control of multiple follower UAVs with respect to the leader UAV.To approximate the unknown nonlinear functions due to the actuator faults and the nonlinearities inherent in the UAV model,FNN approximator is first investigated for controller design.By using the Taylor series expansion,the approximation capacity of FNN approximator is analyzed.Then,by using the basic graph theory,a second-order SME is constructed to estimate the leader UAV’s attitude.Furthermore,with the estimated knowledge,an FNN-based adaptive attitude FOFTCC scheme is proposed for the attitude tracking control of multi-UAVs by using FO sliding-mode surface,FOC,FNN,and adaptive methods.Finally,simulation results have shown the effectiveness of the proposed control scheme.(6)To investigate the safe control of receiver UAVs when they are encountered by the actuator faults and wake vortices induced by the tanker UAV/aircraft in the boom refueling,an adaptive FTCC scheme is developed for the tracking control of receiver UAVs with respect to the tanker UAV/aircraft.By using the functional decomposition,The receiver UAV model is divided into the longitudinal and lateral-directional models.By utilizing the wake vortex modeling technique,the velocities and angular rates induced by the wake vortex are analyzed.Then,FTC schemes are constructed for the longitudinal and lateral-directional subsystems,respectively.Finally,simulation results have shown that all receiver UAVs can safely approach to the predefined positions with respect to the tanker UAV/aircraft even when these receiver UAVs are subjected to the wake vortex and abrupt actuator faults.