Research On The System Design, Dynamic Modeling And Control Methods Of Flying Saucer

Flying saucer is a kind of aircraft using inner rotor system as major power unit. Ithas the ability of VTOL (vertical take-off and landing) and hover, which is similar toa helicopter. Compared with helicopter, flying saucer has special advantages inaerodynamic efficiency and secutity for its built-in rotor system. The research on theflying saucer is still in the development stage, and there are still many problems to besolved about the structure design, aerodynamic optimization, flight control, and so on.The flight performance is unable to meet the needs of practical application, and theconfiguration design, flight control algorithm and other aspects need further researchand development.This dissertation studies and designs a Dual-Rotor Flying Saucer (DRFS), whichhas ducted dual-rotor system and under-mounted pendulum structure. The kinematicsmodel and dynamic model of the aircraft are developed respectively. Based on theanalysis of system characters, the research on the aircraft’s flight control is carried out.The main contributions are as follows:(1) A Maglev Dual-Rotor Flying Saucer SystemThis dissertation illustrates the design of a maglev dual-rotor flying saucer system(DRFS system), whose major structural features includes: the magnetic suspensionmodule is fixed in the end of the rotor, constituting a maglev rotor, and is able toachieve axially positioning the end of the blade; two maglev rotors are fixed up anddown in the duct, which constitute a ducted dual-rotor system, and provode the liftforce and the yawing moment; the cabin is connected with the dual-rotor system bytwo orthogonal joints which mutually vertical to each other, and the moment acting onthe orthogonal joints adjust the attitude of aircraft body. The electronic system of theaircraft is hierarchical architecture: the organization layer has an embeddedIC(Embedded Industrial Computer: EIC), supplemented by a variety ofhuman-machine interface, and it is responsible for monitoring, acquiring information,decision-making and flight control instructions issued; the coordinaton layer has aDSP(Digital Siginal Processor) as the core, supplemented by state sensors, and it ismainly responsible for basic flight control; the execution layer is the motor servosystem, and it is responsible for controlling the motors of the dual-rotor system andthe orthogonal joints. The general parameters of the DRFS include physical systemparameters and aerodynamic parameters. Based on the analysis of DRFS’saerodynamic characters and the choosen physical system parameters, the flyingperformance parameters, such as maximum payload, flying power, maximumclimbing velocity, etc., are calculated.(2) Dynamic Modeling and Analysis of DRFS In this dissertation, the modeling characters of dual-rotor helicopter andducted-fan aircraft are referred, and the aerodynamic model of the ducted dual-rotorsystem is built based on the helicopter blades aerodynamic principles. Furthermore,the dynamic model of DRFS is built based on Kane method, and the dynamic modelis verified by some simulation experiments. The verifying experiments also reveal thehandling and stability characteristics of DRFS: the aircraft has selt-stability with zeroinput; the control response is fast and smooth with zero initial state; the dynamiccharacteristics of longitudinal channel and lateral channel are symmetrical. Finally,the dynamic model of DRFS is analized. The subsystem models of yaw channel,longitudinal channel and lateral channel are built, and the small perturbation linearmodel of DRFS is built, too. Based on the linear model, the controllability,observablity and stability of DRFS in the position-fixed hover status are analized. Theanalysis of controllability and observablity provide foundation and guidance for thedesign of flight controller, and the stability analysis verifies the self-stability of DRFS.(3) Unconstrained Flight Control of DRFSThis dissertation design the flyng control system for DRFS based on the flyingdynamic characters of DRFS and the control requirements. The proper unconstrainedcontrol methods are respectively designed for the linear model, nonlinear model andnonlinear model with parameter disturbance, and accomplish the flight control tasksincluding attitude control, position-fixed hover, trajectory tracking and immunity test.Firstly, the quadratic optimal controller of DRFS is designed based on the smalldisturbance linear model of DRFS, and the linear quadratic regulator (LQR) of aircraftbody is designed according the coupling character of system model. The linearcontroller accomplishes attitude control, position-fixed hover control and trajectorytracking control. Secondly, the dual-loop nonlinear PID control algorithm is designedfor the nonlinear model of DRFS. The attitude control is achieved by inner nonlinearPID controller (Inner NPID), and position and velocity control is achieved by outlinear PID controller (Outer PID). The reference siginals of inner-loop roll and pitchattitude control are produced by Outer PID controller. The simulation verifies therobustness and dynamic performance of the dual-loop nonlinear PID controller, and itshows that the method is feasible and effective for flight control of DRFS. Finally,aiming at solving the parameter’s uncertain change problem caused by aerodynamicdisturbance or payload change, the fault-tolerant adaptive algorithm is designed forthe nonlinear model with parameter’s uncertain change. The stability of the alrorithmis verified based on the analysis of Lyapunov function. Since the on-linear estimationcompensates the bad influence brought by the model parameters’ uncertain change,the stability and control accuracy of the system with parameters’ uncertain change areguaranteed, and this is verified by the simulation results.(4) Research on the Nonlinear Constrained Optimal Control of DRFS This dissertation designs the nonlinear model predictive control (MPC) algorithmfor DRFS based on model predictive control theory. The close-loop optimization ofMPC guarantees the robustness of the algorithm, and the key points of MPC appliedto DRFS are the stability, instantaneity and optimality of the algorithm. Severalalgorithms based on MPC are designed for DRFS. Firstly, considering that thefinite horizon optimal control problem (FHOCP) affects the performance of MPC, anapproximate finite horizon optimal control problem (AFHOCP) is proposed tosubstitute the FHOCP. The method approximately estimates the state variablesequence, and turns the complex nonlinear cost function to a quadratic function. Therelative analysis and experiments show that the method greatly reduces the calculationtime, and keeps the stability of the optimization calculation. Secondly, the serialpredictive decoupling control (SPDC) algorithm is designed according to thehalf-coupling character of DRFS’s nonlinear model, in which MPC controllers arerespectively designed for two sub-systems with serial relation. The serial relation oftwo MPC controllers is built by using the predictive information of the MPCcontrollers. According to the robust MPC theory, the stability of the SPDC algorithmis analized. The simulation result shows that the calculation time is reduced, and theoptimality is improved. Finally, based on the SPDC algorithm, the slide-mode MPC(SMMPC) algorithm is proposed, and its stability is analysized according to theLyapunov stability theory. Piecewise-constant strategy and pre-calculated control areadopted to improve the instantaneity of SMMPC algorithm. The experiments showthat the proposed methods satisfy the requirement of real-time flight control, andachieve expected performance in the stability and optimality.This subject is supported by National Natural Science Foundation of China(No.61075110), China’s863Program (No.2007AA04Z226), Key Project(No.KZ201210005001) of S&T Plan of Beijing Municipal Commission of Education,Beijing Natural Science Foundation (No.4102011), and Specialized Research Fundfor the Doctoral Program of Higher Education (No.20101103110007). The researchresults have significance and reference-value to optimizing the system structure offlying saucer, the analysis of the flying saucer’s dymamic characteristics, and thestudy of flight control problems. Several patents have been granted to the proposedflying saucer, which has application value in the fields of research and education forUAV technology and control science, and military/civilian UAV development.