Kimio Kanai

Design of restructurable flight control systems using feedback linearization

In this paper, a design method is presented for restructurable flight control systems based on the feedback linearization method. Failures are identified indirectly by estimating parameters of the nonlinear aircraft model using the recursive least square algorithm. The aircraft is assumed to have many control surfaces that can be driven independently. In the design, actuator dynamics are taken into account and the Control distributor, which reduces real inputs to generic inputs, is used. The imaginary actuators for generic inputs are introduced to generate input signals used in parameter identification. Pitch and roll angles are controlled indirectly by controlling pitch and roll rates, respectively, which is an approximate way but makes the control system simpler than applying the feedback linearization method straight to the control of the angles. To evaluate the performance of the restructurable flight control system, two failure cases are simulated on the six-degree-of-freedom nonlinear aircraft model.

NEW DESIGN METHOD FOR PULSE-WIDTH MODULATION CONTROL SYSTEMS VIA DIGITAL REDESIGN

A design method is proposed for pulse-width modulation (PWM) control systems, where magnitude of control input is e xed but pulse width is adjustable. In the proposed method, e rst a discrete-time (DT) control system of the amplitude modulation type is designed, and then it is converted into a PWM control system. Pulse width and delay time for PWM input are determined at each sampling time so that the state error between the DT and PWM control systems can be small. The DT controller is determined by digital redesign of a given continuous-time (CT) controller. The digital redesign method previously proposed by the authors is based on the principle of equivalent area. Thus, in spiteof theinherent nonlinearity of PWM input, inasmuch as linear control theory can be employed in design of the baseline DT or CT controller, the PWM control system is obtained without using nonlinear control theory. In addition, because the proposed methods are described in state-space representation, they are applicable to multi-input/multi-output systems. The effectiveness of the methods is illustrated through computer simulation using a e exible spacecraft model.

Application of a new multivariable model-following method to decoupled flight control

A new model-following controller for multivariable, linear, time-invariant systems is described and applied to the decoupled longitudinal control of a control-configured-vehicle-type aircraft. The controller is composed of an input dynamics compensator and a state feedback block. This approach enables decoupling and control of systems that cannot be done by state feedback alone. The key concept is that of system augmentation by utilizing a unimodular matrix to assure the nonsingularity of a control matrix. Two methods of generating the control input are described. In the first, the control input is synthesized by explicitly using the plant state variables. In the second, the input and output of the plant are used.

Digital Redesign of Linear State-Feedback Law via Principle of Equivalent Areas

those obtained for mode 2. The best results were obtained when the signal from SG3 was used in the feedback loop, as in case of CL2 and CL3, cone rming the initial predictions. Note that CL1 was in the case of FC1 limited in gain to guarantee the controller stability. The mostsevere, FC6 was the last to be tested. Because of the accumulative fatigue of the hardware (piezoelectricelements ),several attempts to close the loop with the MIMO controller were unsuccessful. During each attempt, an actuator on the upper group failed, resulting in a system automatic shutdown. The apparent reason was the proximity of these actuators to the most stressed region of the e n,nearthe loadcell.Therefore,a single input /singleoutput (SISO) controller was designed for commanding only the lower group of actuatorstoreducebuffetinginthee rstbendingmodeofthetailduring a closed-loop run at FC6. When feedback from accelerometer A2 and a nominal gain setting to increase chances of success were used, this SISO controller reduced the rms strain at SG3 by approximately 3% in the frequency range of 10 ‐20 Hz. Between 0 and 100 Hz, the rms strain at SG3 was reduced by approximately 2.5%. Conclusions A full-scale aircraft instrumented to reduce buffet loads was tested. The test represents an important step in the development of adaptive smart structures systems. Two groups of actuators consisting of piezoelectric elements distributed over the structure were designed to achieve authority over the e rst and second modes of the vertical e n. Very promising results were obtained in parametric studies using differentsensors in a two input/two output controllerusing the standard time-invariant linear quadratic Gaussian control law design. Based on the most important performance metric, the strain gauge located at the critical point for fatigue, vertical e n buffet attenuation of 57.5% (mode 2) and 33.3% (modes 1 and 2) for the nominal FC were observedduring thetests.Also,attenuationof18.3% (mode2) and 8.7% (modes 1 and 2) were verie ed for the next most severe buffeting case. In general, the CL that included at least one strain gauge in the feedback loop revealed better performance. This is an indication that strain gauges can be better correlated to the control objective, which is to reduce the structural strain generated by buffeting.

Digital redesign methods based on plant input mapping and a new discrete-time model

The digital redesign approach is to obtain discrete-time controllers which realize a digital control system approximating some characteristics of the predesigned continuous-time control system. This paper describes a digital redesign method for linear time-invariant continuous-time control systems of the 1-DOF type and the state-feedback type. It is a closed-loop method that takes account of the closed-loop characteristics of the predesigned continuous-time control system. Generally, such methods provide better closed-loop stability and closer approximation to the underlying continuous-time control system for a wider range of sampling periods. The proposed method is based on mapping of the closed-loop control inputs and the transfer function matrix from external inputs to control inputs is discretized as a discrete-time model which generates average continuous-time system outputs for constant inputs. The discrete-time controller or the feedback/feedforward gain matrix is determined so that the discretized closed-loop transfer function matrix can be realized. The redesign is carried out using the state-space approach; thereby the method can be applicable to MIMO systems. To illustrate the methods effectiveness, numerical examples are shown.

Design of a Pulse-Width-Modulation Spacecraft Attitude Control System Via Digital Redesign

Abstract This paper proposes a design method for pulse-width modulation (PWM) control systems. In the method, a discrete-time (DT) control system of the amplitude modulation type is first designed, and then it is converted into a PWM control system. A pulse width and a delay time for each PWM input are determined based on the principle of equivalent area (PEA) so that the output error between the DT and PWM control systems can be as small as possible. In this paper, the DT controller is determined by the digital redesign of a given continuous-time (CT) controller. The digital redesign method is also based on the PEA. Thus, in spite of the inherent nonlinearity of the PWM input, linear control theories can be employed in the controller design. The effectiveness of the proposed method is illustrated through computer simulation using a linear model of the Japanese HOPE spacecraft

A new digital redesign method for pulse-width modulation control systems

This paper proposes a new design approach to pulsewidth modulation control systems, which redesigns a predesigned linear state feedback law into a pulsewidth modulation control law. The proposed method consists of two steps. The first step is to redesign the predesigned continuous-time control law into a discrete-time one. The proposed digital redesign method is based on the principle of equivalent area. In other words, at each sampling interval, the discrete-time control input is given by the mean value of the control input of the underlying continuous-time closed-loop system. The second step is to convert the discrete-time control input into a pulse-width modulation one; namely, pulse-width and delay time for the input are determined at each sampling interval, which make the error between the state equations of the discrete-time and pulse-width modulation control systems. The obtained pulse width corresponds to the principle of equivalent area again. The proposed method is applied to simple examples to illustrate its effectiveness.

Longitudinal Flight Control System Design for Propulsion Controlled Aircraft

Abstract This paper deals with a problem of speed and flight-path control of a propulsion controlled aircraft (Boeing 747) that has no available primary control surfaces. The performance of the control strategy that takes advantage of the pitchingmoment arm difference between the inboard and outboard engines is investigated through computer simulation. Then assuming that the stabilizer is available as a feedforward or feedback input, it is shown that the control becomes easier. It is also shown that considering the variation of the flight condition it is possible to decrease both airspeed and altitude by controlling the flight path angle with thrust only.

Reply by the Authors to Franco Bernelli-Zazzera and Paolo Mantegazza

References 1Ieko, T., Ochi, Y., and Kanai, K., “New Design Method for Pulse-Width Modulation Control Systems via Digital Redesign,” Journal of Guidance, Control, and Dynamics, Vol. 22, No. 1, 1999, pp. 123–128. 2Bernelli-Zazzera, F., and Mantegazza, P., “Pulse-Width Equivalent to Pulse-AmplitudeDiscrete Controlof Linear Systems,” Journal ofGuidance, Control, and Dynamics, Vol. 15, No. 2, 1992, pp. 461–467. 3Bernelli-Zazzera, F., and Mantegazza, P., “Linearization Techniques for Pulse Width Control of Linear Systems,” Control and Dynamic Systems, Vol. 70, Academic Press, New York, 1995, pp. 67–111. 4Bernelli-Zazzera, F., and Mantegazza, P., “Control of Flexible Structures by Means of Air Jet Thrusters: Experimental Results,” Proceedings of the Ninth Virginia Polytechnic Institute and State University Symposium on Dynamics and Control of Large Structures, edited by L. Meirovitch, Blacksburg, Virginia, May 1993, pp. 231–242. 5Shieh,L., Wang,W., and Sunkel, J. W., “Design of PAM and PWMControllers for Sampled-Data Interval Systems,” Journal of Dynamic Systems, Measurement, and Control, Vol. 118, No. 4, 1996, pp. 673–682. 6Bernelli-Zazzera, F.,Mantegazza, P., and Nurzia, V., “Multi Pulse-Width Modulated Control of Linear Systems,” Journal of Guidance, Control, and Dynamics, Vol. 21, No. 1, 1998, pp. 64–70.

Modeling of Aerodynamic Uncertainties and Introduction of Dynamic Pressure Compensation to Robust Flight Control Systems.

This paper first describes a modeling method for the uncertainty of aerodynamic coefficients in a linear state-space aircraft model. Second, a design method of robust flight control systems based on the model is also presented. The proposed method is applied to the structure of the uncertainty for the μ-analysis and has the following characteristics: 1) the variations of the altitude and the velocity are explicitly treated as real repeated scalar perturbations since all dimensional derivatives are the function of the altitude and the velocity, and as a result 2) a gain-scheduled controller using the dynamic pressure compensation which satisfies the robust performance for a comparatively wider flight envelope can be designed. Simulation study is carried out to indicate that the proposed method is less conservative and satisfies the robust performance for a wider flight envelope than those of the H∞ design and the μ-synthesis where all aerodynamic coefficients are assumed to be individually perturbed.