Search

Judder is well known as a vibration that occurs in a dry clutch. In order to maintain the ride quality and quietness of dry clutches, research on judder analysis and mechanical design to suppress judder is actively pursued. This paper describes the design and implementation of robust control techniques as an approach to suppressing judder during gear shifting in traction-controlled transmissions. This paper is concerned with transmissions that exhibit strongly nonlinear frictional characteristics that can only be approximated by higher order polynomials. Specifically, the coefficient of friction varies significantly nonlinearly with the rotational speed during gear shifting. In this paper, the controller is designed by expressing this nonlinear property as the uncertainty of the frequency. Simulation and bench test results are presented to validate the validity of such an approach. The results show that judder-free shifting is possible with the proposed approach. Furthermore, the requirements of three controllers are presented by analyzing the robust controllers.

This paper presents a novel robust control approach for a fully-actuated hexrotor based on Gaussian process regression(GPR). A state-space model considering aerodynamic disturbances as unknown dynamics is first built as the hexrotor dynamical model. GPR is next introduced as a learning method of the unknown dynamics. Then, a robust control method to ensure uniform ultimate boundedness on the control error is proposed. Finally, the effectiveness of the present method is shown by landing experiments near a wall.

This paper provides comprehensive design method of compensator for grid-connected PWM inverters with LCL-filter to supply high quality electric power from distributed generations. The proposed compensator ensures: i) high speed tracking with low distortion and zero steady-state errors for sinusoidal current reference; ii) robust responses against frequency and voltage changes of grid power supply. The proposed design methodology is based on the utmost compensator that covers all realizable linear control systems. Various multi-objective control requirements can be considered in the design framework directly, and then, trial-and-error process is reduced. The software-in-the-loop simulations are presented to support the advantages of the design and to demonstrate the system performance compared to traditional Linear Quadratic Regulator with Internal Model controller., 本文データは計測自動制御学会の許諾に基づきJ-STAGEから複製したものである.https://www.jstage.jst.go.jp/article/sicetr/51/3/51_162/_article/-char/ja/

This paper presents an application of the model predictive control to flexible spacecraft attitude maneuver. The objective of the study is to demonstrate an experimental feasibility of optimal attitude controllers suppressing the flexible appendage vibration under the actuator saturation. For the purpose, we study two approaches in the model predictive control framework. One is to use the frequency-shaped linear quadratic cost with finite horizon, and the other is to apply the robust control theory. Through ground experiments using a single axis air bearing table, as well as numerical studies for ETS-VIII spacecraft, their abilities are shown and compared., 資料番号: PA1420038000

制御系解析・設計における数値計算/数式処理ソフトウェアの活用

In this report, a robust controller for underwater robots equipped with a manipulator (underwater robot manipulators) is developed to overcome uncertainties in robot dynamics and hydrodynamic forces. The proposed controller has the following features : (1) the structure of the controller is very simple in comparison with that of an adaptive controller using a regressor of a robot dynamic equation； (2) the controller is developed without using a function specific to sliding mode type robust control, such as a signum function and a saturation function, which often leads to oscillations in control inputs； and (3) the maximum value of a tracking error norm can be arbitrarily reduced by setting controller design parameters. Furthermore, numerical simulations whose model is a 2 link planar underwater robot manipulator are performed. The simulation results demonstrate the validity of the theoretical results (the controller features).

2, KJ00005988341, 論文, Article

In this paper, a method of automatic controller design for electronic control systems is described. In order to automate the design of an electronic controller, an evolutionary computation is applied. First, the framework for applying the genetic algorithm to the automation of controller design is described. In particular, the coding of a chromosome is shown in detail. Then, how to make a fitness function is represented, with an air conditioner as an example, and the controller of the air conditioner is developed automatically using our proposed framework. Finally, an evolutionary simulation is performed to confirm our framework.

In this report, we confirm the effectiveness of the proposed multirate adaptive output feedback control method in the 1st report through numerical simulations and experiments for a cart-crane system. The cart-crane system is typical non-square system with single-input and two-outputs. The proposed adaptive method allows us to apply the almost strictly positive real(ASPR) based adaptive output feedback scheme with the dual-rate sampling strategy for the control of the cart-crane system.

This paper presents a design method of a switching control algorithm in which the conservativeness is reduced in sliding-mode control. In a conventional sliding-mode control system, the control input is switched so that a certain function can remain a Lyapunov one in spite of the presence of system uncertainties. However, the switching gain with conservativeness is generally derived due to the simplicity of the control algorithm. In this case, higher gain is required than is necessary. This situation is not suitable since it causes chattering in system inputs and/or steady state errors in system outputs. In this paper, a condition realizing a Lyapunov function is newly derived. Based on this result, a novel switching control algorithm with reduced conservativeness is proposed. The algorithm is also expanded to an output-tracking control problem. Finally, simulation results are given to illustrate the effectiveness of the proposed switching control algorithm.

In adaptive output feedback control based on almost strictly positive real conditions, a technical difficulty arise when the multi-input, multi-output (MIMO) system under consideration is nonsquare. To overcome this problem, the idea of multirate sampled-data control will be proposed. That is, through careful choice of faster input sampling rates, we create a lifted discrete-time system which has the same number of inputs and outputs without the causality constraint. The output feedback based adaptive control strategy can then be applied to this lifted system under certain conditions. In this report, we also propose a robust adaptive output feedback controller design scheme for non-square MIMO systems using the multirate sampling strategy which can solve the causality problem and show that all the signals in the resulting closed loop control system are bounded.

資料番号: SA1150246000

By conventional study of predictive control for multiple input time delay system, state integral transformation was shown, but control problem has not been investigated in detail. The author has considered about robust predictive control based on the H∞ control theory for multiple input time delay system, and has proposed a problem setting with multiple structural model errors, a description of control problem, a control system design method using scaling matrix. But some problems exist about description of generalized plant when multiple model errors of time delays is considered. In this paper, a description of generalized plant of multiple input time delay system for robust control is proposed.

This paper presents a nonlinear controller for electric power assist systems based on the sliding mode control. The proposed sliding mode controller is designed to achieve desired nonlinear properties including gravitational effect with robustness against disturbances such as friction force and modeling errors. Furthermore, so-called reaching phase is positively utilized so that the operator can feel disturbance torque of relatively large amplitude which should be noticed as information about environment, for example, when hitting an obstacle accidentally. Since the reaching phase should be designed from several points of view, the dynamics in the reaching phase is linearlized in order to use fruitful linear control theories such as H_∞ theory.

This paper presents the design method of a preview-sliding-mode controller for discrete-time linear invariant systems with disturbance. As well as the integral sliding mode controller in continuous time, the control law consists of two components; a linear component for the nominal system, and a variable structure component to cancel the distubance. Since the linear component is designed by means of optimal preview control technique, the closed loop system achieves optimal tracking performance robustly.

In this paper, flight controller design for the lateral/direction motions of in-flight simulator MuPAL-α is discussed. For the simultaneous simulation of gust responses and pilot input responses of model aircraft, feedback and feed forward controllers are designed in two steps; feedback controllers for robust model-matching problem are first designed using scaled H_∞ control to simulate the gust responses of target aircraft, and then feed forward controllers for H_∞, model-matching are designed as right inverse systems to simulate the aircraft responses driven by pilot inputs. In the feedback controller design, we use airspeed in stead of inertial speed which was used in our previous report. After confirming that designed controllers have good model-matching performance with numerical simulations, we conducted hardware-in-the-loop simulations and flight tests. In the flight tests, we confirmed that controllers with air speed feedback have better model-matching performance than controllers with inertial speed feedback, i.e. we have improved the model-matching performance of in-flight simulator MuPAL-α for its lateral/directional motions.

In Japan, since the development of high density and large-scale power system, the multimode oscillations occurs in the generator connected with this the modern electric power system. There have been mixed mode between the power systems and many others oscillations, such as the SSR and cross-compound machine oscillations, which is specific mode oscillations of generator. To date, most major generators in trunk power system are equipped with supplementary excitation control, commonly referred to as the conventional PSS. However, low frequency oscillations still occur in system. In this paper, first the development of adaptive H_∞ control system for stability and robust control is explained. Then, a goal for control design is verified by simulation on the actual model with new adaptive H_∞ control system.

When a railway vehicle negotiates curved tracks at high-speeds, the car body collides with lateral bump/stop in the secondary suspension due to excess centrifugal force, and the ride comfort gets worse. In this situation, there is also a possibility that active suspension does not work effectively. This paper deals with the addition of centering control, which pulls the car body back to the central direction from the outer side, to the vibration control in the active suspension. Considering the difference in frequency range to be controlled between the vibration and centering controls, H_∞ control law is employed and an effective weighting function is offered for cooperation of the two controls. Numerical simulations show that cooperation can be realized in spite of the trade-off relation between the two controls. The cooperative work of both controls is confirmed in the experiment using a scaled half vehicle model under the condition with equivalent excess centrifugal force.

Nonlinear control methods for automotive electronic throttle control (ETC) systems are developed using the theory of sliding mode control. The motivation for the use of sliding mode controllers stems from the fact that the throttle valve in our developing ETC is supported for fail-safe by two springs which can be modeled as a nonlinear discontinuous spring. This situation, however, satisfies so called matching condition in the context of the sliding mode control theory, namely the system is insensitive to this disturbance during sliding mode. Two kind of methods to estimate the valve velocity with the measurement of its position, variable structure observer and over-sampling method, are also designed. Experimental results illustrate the effectiveness of the controller and estimators.

This paper presents a new sliding mode controller based on the theory of model following control for semi-active suspension systems with Magnetorheological (MR) dampers which have undesirable nonlinear properties. In the proposed controller, a desired semi-active suspension system is chosen as the reference model to be followed, and the control low is determined so that an asymptotically stable sliding mode will occur in the error dynamics between the plant and the reference model states. The advantages of the proposed controller are : (1) measurement of the damping force is not required, (2) the reference model specifies the desired performance considering the passivity constraint of the damper, (3) it is highly possible to maintain the sliding mode and achieve high robustness against the nonlinear properties of the damper. Numerical simulations illustrate the effectiveness of the controller.

This paper proposes Model Predictive Control with I-compensator of PID control to a traveling crane system. I-element improves disturbance removal characteristics of Model Predictive Control. To verify the control performance of the proposed method, we compared the proposed method with former one by swing suppression and positioning control experiment of a traveling crane system. As a result, in the step disturbance case, the proposed method can control the system without offset, but former one can't do. And the control performance of the proposed method under the sinusoid disturbance is superior to former one., 論文

The fundamental properties of active magnetic bearing controllers are investigated from the standpoint of output regulation in the presence of deterministic disturbances. A transfer function approach is used in the analysis. The variable to be regulated is selected from rotor displacement, coil current and bearing force. The treated disturbance is constant or harmonic ; the latter is caused by rotor unbalance. The essential properties of the controller achieving output regulation for constant disturbances are derived. Those for unbalance disturbance are also shown by using complexvariable representation, which makes the procedure of analysis similar to that for constant disturbances. The derived controllers for unbalance compensation are converted into two types of realvariable representation : state equation and convolution integral. It clarifies the equivalence between the independently developed unbalance compensation controllers. It is also shown that the obtained results can be utilized in designing controllers.

宇宙構造物の制御の問題は1970年代後半から活発な研究が行われてきた。特にロバスト制御の適用について多くの議論がなされ、様々な地上実験も試みられてきている。しかし、実衛星を用いた軌道上での実証実験については、その重要性にもかかわらず報告されていない。軌道上実験を行う際に特に問題となることは、実機の軌道上での柔軟パラメータの推定が困難であり、軌道上モデルの精度に限界があることと、搭載制御系のハードウェアの能力やメモリの容量の制約から弾性振動モードのうち制御可能なモード数に限界があることである。このため、技術試験衛星6型(ETS-6)による軌道上実験では、打ち上げ前の有限要索法や部分構造合成法などによる地上同定と軌道上でのガスジェットスラスタ加振による軌道上同定の比較をまずはじめに行い、その結果を制御系設計に反映させ、より高精度な振動・姿勢制御実験を行うことが可能になった。本報告ではこれらの軌道上制御実験で適用された、周波数成形LQG(線形2次ガウス)制御やH(sub ∞)制御などのいわゆるロバスト制御に基づいた制御系設計法について、そのアルゴリズムの概要と軌道上実験結果についてまとめる。なお、ロバスト制御理論による制御系設計法に関する研究は、航空宇宙技術研究所と株式会社東芝との共同研究結果によるものであり、またETS-6による軌道上制御実験はその成果を実証するために、航空宇宙技術研究所と宇宙開発事業団との共同研究の中で実施されたものである。, Engineering Test Satellite-6 (ETS-6) is a three axis stabilized satellite that was launched by NASDA in 1994. The Flexible structure modeling and control experiments project (FLEX), which consist of identification and attitude control experiments on orbit, were planned as one of the ETS-6 onboard experimental programs and carried out in 1995. The purpose of the FLEX experiments was to establish fundamental attitude controller design methods for a flexible structure satellite. In this report, some results of the attitude control experiments are presented. The attitude controllers are designed using LQG (Linear Quadratic Gaussian) and H(sub infinity) control methods. Experimental results have shown high control performance as expected by simulation results on the ground., 資料番号: AA0001662000, レポート番号: NAL TR-1348

In recent years, active suspensions have been studied for improving the ride quality of railway vehicles. In the active vibration control, it is important that the control system be robust against mass variation of the car body due to the varying number of passengers. However, it was indicated that LQG controller can become unstable when the body mass is lightened. Although H_∞ controller is stable against the mass variation, it is expected that the adoption of μ synthesis can improve the control performance still more. In this paper, the car body mass variation is considered in the linear state space representation of control object as parametric uncertainty, so that the μ synthesis can be adopted for the controller design. The control performance of the μ controller is investigated with a 3 d. o. f. half-vehicle model experimentally. As a result, it is shown that the μ controller improves the deterioration due to the car body mass reduction, and that it provides better control performance using a frequency dependent weighting function more than using constant weightings for the uncertainty of the mass variation.

In this paper, the equations of motion of the non-contact hold and transfer control system of a spherical magnetic body using a one-pole-type magnetic robot hand system are derived. The robot hand system consists of a two-link robot arm and a one-pole-type magnetic hand. The Iinearized state equation and the linearized output equation of the magnetic robot hand, are derived and the controllability and observability are then clarihed. The non-contact hold control system by the magnetic robot hand is then constructed using H_∞ control based on the normalized coprime factorization method, and the tracking control system of the robot arm is constructed using the disturbance observer. It is confirmed from the experimental results that the non-contact hold and transfer control of the spherical magnetic body can be implemented well using the present control system.

Current control algorithms for antilock braking system (ABS) are mostly table-driven and should be frequently modified through various experiments, because the braking dynamics are highly nonlinear with system uncertaintieS. In this paper, a new control algorithm is proposed for ABS using the theory of sliding mode control based on a linearized system model, which represents nonlinearities as external disturbances. The proposed sliding mode controller guarantees a highly robust performance against large variations in the system parameters and disturbances. Furthermore, a road surface estimation scheme using a disturbance observer is proposed in order to determine the optimal reference input. Simulation results reveal the advantages of the proposed method in comparison with conventional linear controllers.

As nonlinearity and complexity of a nonlinear system increase, it may be more difficult to construct a controller by the mathematical control theory. In such cases, it is very effective to construct the controller by using Neural Network(NN), because NNs have capabilities of coping with the nonlinearity and complexity of the nonlinear systems. NN controllers are constructed through learning to minimize a criterion function under certain circumstances. But NN controllers may not work well under very different circumstances from those at learning stage. For example, NN controllers are usually made without considering disturbances because NN controllers do not have a means to supress their influences. So, in case of existing the disturbances NN controllers do not work well. In this paper a robust control system design method for the disturbances is discussed using second order derivatives of Universal Learning Network.

In this paper, we propose a new simple design technique for constructing digital PID compensators for the robust control of sampled-data systems, based on a minimax procedure. First, we formulate the robust control problems for sampled-data systems, including digital PID compensators and disturbances of system parameters. Next, we propose the minimax-based design method with the PID compensators for the robust control problems of sampled-data systems, where we develop a new minimax algorithm which is reliable, stable and simple in computation. Finally, we apply the proposed method to a realistic problem in order to demonstrate its effectiveness.

Recently, the application of active suspension has been studied for improving the ride quality of railway vehicles. In active vibration control of railway vehicles, it is important that the control system be robust against parameter variations. In one such study, it was indicated that a LQG controller had low stability of the robustness to mass variation of the car body due to the varying number of passengers. In this paper, the H_∞ controllaw is adopted for a 3 d. o. f. half-vehicle model, and the improvement of control performance is investigated experimentally. As a result, it is shown that the addition of a bogie-related state variable to body-related ones provides better control performance. The control performance is improved further by shortening the sampling period. Moreover, it is also shown that the H_∞ controller is more robust against mass variation of the car body.

This paper presents the local control approach for a 1/2-scale controlled-PM LSM Maglev test vehicle in our Laboratory. Under an extremely unbalanced condition of one controlled-PM of which a coercive force is reduced strongly, the 256 kg-heavy-vehicle has been levitated in a dynamically stable state with compensating slowly bending motions of the beams connecting the four controlled-PM’s. Minimum power-loss control method for stationary levitation mode and constant-gap-control method for Maglev-starting and -stopping mode have been carried out by using an optimum robust control theory of the digital feedback levitation system. Experimental system is discussed and the experiment results are presented. It is also shown that the control system has provided minimum power-loss levitation and satisfied constant-gap control without noise.

In this paper, we propose a systematic controller design which treats problems of system identification and robust controller design, for a nonlinear heat-transfer process. First, we construct a convenient parametric model of the object possessing the nonlinear characteristic. Then, by using the modelling error between the real system and the model as the dominant design parameter, we construct a robust control system based on a mixed-sensitivity approach. This method avoids the need to linearize of a nonlinear system and provides robustness against nonlinear input. The effectiveness of the control method is verified by both numerical simulation and direct experiment for producing a constantly rising temperature of a microheater., 論文

The aim of this paper is to propose a new method for the optimal "H_∞ norm" computation of time-varying systems. When we define the L_2-induced norm as "H- norm", the optimal "H_∞ norm" computation problem can be formulated as a minimax problem involving control and exogenous inputs. Our approach is as follows. First, we derive analytical solutions for the minimax problems with the exogenous inputs being frozen. Secondly, we present an algorithm for numerical solutions of the minimax problem with the control being frozen : Here, we prove that accumulation points generated by the algorithm satisfy the necessary conditions for optimality. Thirdly, combining the two above-mentioned techniques, we propose a relatively simple method for minimax solutions under assumptions that the saddle points exist. Lastly, we demonstrate the effectiveness of our approach by some simulations.

リフティングボディ形状の機体は，宇宙往還機の候補の一つとして考えられている．翼胴形状の航空機に比べて再突入時の空力加熱に比較的強い構造であること，より大きなペイロード容積を確保できること，ロケット先端のフェアリング内に収まりやすい形状であることなど有利な点が多い．揚力を利用して飛行するため，カプセル形状に比べればより柔軟な飛行制御が可能となる．その一方で，揚抗比が極めて小さく，飛行性能は通常の航空機に比べ劣り，低速での飛行制御が困難である．特に滑走路への着陸では，低速であるが精度の高い飛行制御が要求される．この技術課題の克服に向け，小型模型を用いたリフティングボディ飛行実験(LIFLEX)が計画された．飛行実証においては，着陸性能を確保する飛行制御系が技術開発の中心となる．実際に想定される外乱や機体モデル誤差などの様々な不確定要因の存在下において，求められる着陸性能を確保する必要がある．本稿では誘導制御系設計の概要について触れた後に，数値シミュレーションによるシステム評価と設計の改善について述べる．まず様々な不確定要素を組み込んだシステムを，数値シミュレーションにより評価した．次に，非線形システムの設計パラメタを直接最適化するために開発した手法を用い，不確定性に対するロバスト性を改善した．さらに今後の開発の参考資料とするため，小型模型実験機である本実験機固有の設計条件を見直し，より一般のシステムに適用可能な条件を用いて，システムを評価した．, The lifting-body shaped vehicle has been studied as the candidate for a future re-entry vehicle. Because of a wingless structure, there are several advantages, such as structural stiffness against atmospheric heating, greater payload capacity, and suitable shape stored in a rocket fairing. Furthermore, since it flies utilizing aerodynamic force, the vehicle motion can be controlled more flexible compared to a capsule-type re-entry vehicle. However, the controllability is inferior to a normal aircraft and flight control becomes difficult at low air speed because of a low lift-drag ratio. Especially in approach and landing phase, accurate flight control is required while the air speed is quite low. To establish landing technology for a lifting-body flight vehicle, LIfting-body FLight EXperiment (LIFLEX) had been planned. In this experiment, the key technology to be developed is a flight control system. The experimental vehicle has to meet given specifications and land on a run-way against various uncertain conditions such as environmental uncertainties and vehicle model errors which are expected to exist in the real world. This paper describes an overview of the control system design for LIFLEX, and system evaluation by a numerical flight simulation. First, the system performance and robustness is evaluated incorporating those uncertain parameters. Then, the control design parameters with the non-linear system are directly optimized to maximize a probability of mission achievement. Finally, uncertain parameters, which are inherent for the LIFLEX vehicle, are revised for more general flight system, and the revised system is evaluated to confirm its robustness as additional information for the development of a future re-entry vehicle., 形態: カラー図版あり, Physical characteristics: Original contains color illustrations, 資料番号: AA0064985000, レポート番号: JAXA-RR-10-007

Thermoluminescence dating has attracted the attention of geology and archaeology researchers as a new method. This method requires a heating device to raise the temperature of the sample continuously at an exact rate. This paper describes the design method of a control system which can accomplish this goal. This control system tends to be unstable because of the existence of parameter perturbation and lag time. In this paper the perturbation and the lag time are treated as perturbations of a generalized system regarded as a one-order system. Tracking characteristics in the low-frequency domain and robust stability in the high-frequency domain of the control system are achieved using the mixed sensitivity problem which is a kind of H^∞ control theory. Experiments verify the effectiveness of this method., 論文