In this paper, we discuss the influence which changed by the music for humans. We propose measurement of the EEG and feature extraction by the multivariate analysis. Then, we try to classify by the neural network in orderto verify the features. We compare accuracy of the features by recognition rate

View full abstract

[in Japanese]

View full abstract

Compressed digital audio technology removes high frequency component data, which causes sound quality degradation. We have applied Sampled-Data Control to high frequency compensation of compressed audio signals. As a result, high frequency components are compensated so that they are close to the frequency spectrum of CDs. PEAQ(Perceived Evaluation of Audio Quality) also shows that the sound is remarkably improved.

View full abstract

The case of the discrete data analysis based on the multiresolution analysis is examined as a concrete algorithm of the integral wavelet transform at present time. On the other hand, authors previously proposed a new method of the integral wavelet transform. Here, we report on the decomposition and reconstruction of a non-stationary signal in the proposed method.

View full abstract

[in Japanese]

View full abstract

[in Japanese]

View full abstract

[in Japanese]

View full abstract

[in Japanese]

View full abstract

[in Japanese]

View full abstract

[in Japanese]

View full abstract

[in Japanese]

View full abstract

[in Japanese]

View full abstract

[in Japanese]

View full abstract

[in Japanese]

View full abstract

[in Japanese]

View full abstract

[in Japanese]

View full abstract

[in Japanese]

View full abstract

[in Japanese]

View full abstract

[in Japanese]

View full abstract

This paper proposes a new algorithm for acoustic echo cancellation. To separate the near-end signal from the echo, we apply an independent component analysis (ICA) technique to derive a new adaptive algorithm under the assumption that both signals are independent. The local convergence condition is also shown.

View full abstract

[in Japanese]

View full abstract

[in Japanese]

View full abstract

This paper considers time-varying uncertain systems, and develops a method for computing a probabilistic output admissible set, which is a set of initial states probabilistically assured to satisfy the constraint. We have already presented a method for the time-invariant case and this paper shows that, the time-varying case is reduced to the time-invariant case, as long as the constraint horizon is available.

View full abstract

In this paper, an iterative PID controller design method based on the unfalsification of the input-output data and L2 gain criteria is considered. The proposed design method is plant model free and includes the procedure for the adjustment of the performance specification and the specification is updataed by SVM. The effectiveness of the proposed method is evaluated by experiments on a magnetic levitation system.

View full abstract

[in Japanese]

View full abstract

[in Japanese]

View full abstract

[in Japanese]

View full abstract

[in Japanese]

View full abstract

[in Japanese]

View full abstract

[in Japanese]

View full abstract

In this paper, we propose a global finite-time stabilizer for a wheeled mobile robot.Moreover, we analyze the oscillation of the controlled system using homogeneous eigenvalue analysis.Finally, we confirm the effectiveness of the proposed method by computer simulation.

View full abstract

[in Japanese]

View full abstract

[in Japanese]

View full abstract

[in Japanese]

View full abstract

[in Japanese]

View full abstract

[in Japanese]

View full abstract

[in Japanese]

View full abstract

[in Japanese]

View full abstract

This paper considers global stability conditions for discrete-time complex-valued recurrent neural networks, which are regarded as nonlinear dynamical feedback systems. A globally asymptotic stability condition for the networks is derived by way of a suitable choice of activation functions. The condition is shown to be successfully applied to a convex optimization problem, for which real field solution methods are generally tedious.

View full abstract

This paper proposes a max-plus model predictive control (MPC) strategy fordiscrete event dynamical systems modelled by max-plus-linear systemswith interval parameters.The control law will be derived in the form of a range of control input vector when the system parameters of max-plus-linear systems varies within the given intervals.

View full abstract

[in Japanese]

View full abstract

We develop dynamics of a space robot which has a hyperregular Lagrangian in the context of Lagrangian systems. First, we show how the symplectic structure can be induced on the tangent bundle by using the Legendre transformation. Then, we formulate an intrinsic Lagrangian system with external forces and demonstrate an example of 2-dimensional space robots to derive equations of motion with conservations of linear and angular momentum.

View full abstract

[in Japanese]

View full abstract

[in Japanese]

View full abstract

Many researches of active steering systems have been proceedingto improve the driving safety.In this research, passivity-based adaptivenonlinear control is proposed for active steering systems.The designed controller is robust to uncertainties and nonlinearities of tire's lateral force.Furthermore, the controller is much simpler than the existing ones.

View full abstract

In this paper, it is considered the influence of the parameter error included in the coordinate transformation which changes a nonholonomic driftless system into a canonical form called ``Chained Form''. And an adaptive control method for a class of nonholonomic driftless systems with an unknown system parameter via Time State Control Form is proposed.

View full abstract

This paper proposes a new method of regional stability analysis of nonlinear systems based on Rantzer's density functions, We show a sufficient condition in terms of density functions to guarantee positive invariance of a set and convergence of almost all trajectories therein. This condition leads to a convex formulation of synthesis with state and input constraints.

View full abstract

In recent years, there are many researches to control nonholonomic systems, because it is said to be difficult to control.In this paper, the lower bound of L1 gain of chained form systems is calcurated and analyzed. And these systems are shown to be difficult to control with disturbance.

View full abstract

[in Japanese]

View full abstract

[in Japanese]

View full abstract