Shigeru Hanba

New semi-active suspension controller design using quasi-linearization and frequency shaping

Abstract A semi-active suspension system is considered to be one of the most realistic solutions for improving the ride quality of large commercial vehicles, because of its smaller energy consumption compared with that of active suspension systems. A semi-active suspension system is classified as a kind of bilinear system to which linear control theory cannot be applied. A new method of designing a controller for semi-active suspension systems using quasi-linearization and frequency shaping is proposed here. The validity of the proposed method is shown by computer simulation results and vehicle test results for a sight-seeing bus.

Brief Output feedback stabilization of bilinear systems using dead-beat observers

An output-feedback stabilizing controller for bilinear systems is proposed in this paper. Key techniques in the proposed method are periodic switching of the controller and the use of a dead-beat observer. The controller is proved to globally stabilize the plant, both in noiseless and noisy conditions. The only condition on the plant, for the proposed method to be applicable, is that the linear part of it is observable. No condition on nonlinear observability is required.

Further Results on the Uniform Observability of Discrete-Time Nonlinear Systems

In this technical note, the relation between notions of uniform observability of discrete-time nonlinear systems based on injectivity of an observation map (window), the full-rankness of its Jacobian, and a K-function is investigated. It is proven that a system is uniformly observable in the sense of injectivity of the observation map (together with the full-rankness of its Jacobian) if and only if it is so in some K-function senses.

A method of damping transient rush voltage fluctuation of power systems due to a wind driven generator via nonlinear state feedback control

Squirrel-cage induction generators are widely used as generators for windmill power systems, because they are inexpensive, of high durability, and capable of operating asynchronously to power systems. However, induction generators have the drawback that it cause the transient rush current several times as large as the rated value when it is connected to power systems. Also,since wind energy is influenced by geographic and weather conditions, start/stop operations occur frequently for windmill power systems. Therefore, connecting induction generators to power systems cause undesirable voltage vibration in power systems, which is the major obstacle against the practical use of the windmill generator systems. This motivates the study of transient rush current/voltage attenuation in windmill generator systems. In this paper, a nonlinear state-feedback controller for windmill power systems is proposed. The proposed controller not only globally asymptotically stabilizes the plant, but also effectively attenuate the transient current/voltage of those systems. Computer simulations using parameters of the actual windmill generator system also give good results.

Adaptive Control for Nonlinear Systems with Unknown Degrees and Uncertain Relative Degrees

Abstract This paper presents a novel approach to the design of adaptive stabilizing controllers for nonlinear systems with unknown degrees and uncertain relative degrees. Contrary to the previous researches where the degrees and relative degrees of systems are known, the degrees in the present paper are completely unknown, and the relative degrees are partly unknown; that is, the relative degrees are known to be r , r + 1, or r + 2 with known r . It is shown that the proposed adaptive controllers stabilize nonlinear systems of unknown degrees without exact knowledge of relative degrees, and that the outputs converge to zero asymptotically.

Robust Nonlinear Model Predictive Control With Variable Block Length

Robust model predictive controllers for discrete-time nonlinear systems are proposed in this technical note. The algorithms first search for an open-loop controller with some block length, and then try to improve it in a closed-loop fashion by solving minimax problems on-line. It is proved that the controllers are capable of making a subsequence of the state converge into a target set in the presence of bounded disturbances.

Controllability to the origin implies state-feedback stabilizability for discrete-time nonlinear systems☆

Abstract The problem of finite-time state-feedback stabilizability of discrete-time nonlinear systems has been considered in this technical communique. Two assertions have been proved. First, if the system is N -step controllable to the origin, then there is a state feedback control law for which the trajectory of the closed-loop system converges to the origin in N steps. Second, if the system is asymptotically controllable to the origin and satisfies the controllability rank condition at the origin, then there is a state feedback control law for which the trajectory of the closed-loop system converges to the origin in finite steps.

Existence of an observation window of finite width for continuous-time autonomous nonlinear systems

In this note, the relationship between notions of observability for continuous-time nonlinear system related to distinguishability, observability rank condition and K-function has been investigated. It is proved that an autonomous nonlinear system that is observable in both distinguishability and rank condition sense permits an observation window of finite width, and it is possible to construct a K-function related to observability for such system.

Globally convergent Jacobian-free nonlinear equation solvers based on non-monotone norm descent conditions and a modified line search technique

A Jacobian-free nonlinear equation solver based on a search technique called a ‘spiral search’, which is a modification of the line search, is proposed in this paper. Under mild conditions, the solver is proved to be globally convergent. The method is then extended to an overdetermined system of nonlinear equations. Numerical results show that for some problems, the solver outperforms existing solvers based on the line search or the trust-region method.

Observers, Adaptive Observers and Output Feedback Stabilizing Controllers for Uniformly Observable Bilinear Systems

Abstract An adaptive observer for uniformly observable bilinear systems is proposed in this paper. The adaptive observer is derived from the generalized observer canonical form proposed by Williamson, where the conventional input- and output-injection are replaced by the injection of the output and the derivatives of the input. Then, an output-feedback stabilizing controller, using the observer dependent on the derivatives of the input, is proposed. To avoid the differentiation of the input, integrators are placed at the input channel, which makes the direct design of the controller impossible. To overcome the problem, an indirect design scheme based on backstepping is proposed.