This paper investigates a networked bilateral control system affected by a contact force. The proposed system consists of rigid master and nonlinear flexible slave arms, and the network which causes a random delay connects both arms. The slave arm collides with an obstacle during its motion. An extended Kalman filter is designed to reduces the effect of the random delay. Numerical simulations is demonstrated to confirm the performance of the proposed system.

For discrete-time linear stochastic systems with unknown disturbances, we consider the optimal filter with disturbance decoupling property and the equation (i.e., Riccati equation) which the covariance matrices of the estimation errors of the filter satisfy. Assuming that the stochastic processes have constant coefficients, we prove convergence of the Riccati equation and derive a simple equation (called the algebraic Riccati equation (ARE)) which is the limit of the Riccati equation under some conditions similar to those for the Kalman filter. We also prove asymptotic stability of the systems whose optimal gains are determined by the ARE.

In this study, we propose a simple probability density function (PDF) model of Gaussian-Laplacian mixture type, which is capable of parameterizing a heavy-tailed data easily. We construct our model of a convex combination of Gaussian and Laplacian PDFs and derive a minimal parameterization of it. Next, we conduct least-squares fitting of our model to a heavy-tailed data generated by a random Duffing oscillator and obtain over 94% of residual sum of squares (RSS) fitness. The resulting model is applied to predicting transient moment responses and yields over 97% of RSS fitness to Monte{Carlo simulation results of the original system.

For the estimation of deer population dynamics, Bayesian method using Gibbs sampler which is one of the Markov Chain Monte Carlo (MCMC) method has been adopted in the previous studies although the method takes much time for calculation. In this study, we introduce Hamiltonian Monte Carlo (HMC) method as a new approach which is more efficient than Gibbs sampling. To verify our proposal method contributes to deer population managements, we modified the model of a previous study to apply Hamiltonian dynamics and conducted numerical experiments by comparing the results which are estimated in two methods with using simulation based data. As a result, we confirmed HMC method is more time-efficient than Gibbs sampler because the number of samples abandoned for after procedures in HMC method was much less than in Gibbs sampler. Therefore, we conclude that the estimation using HMC is valuable for the estimation of deer population dynamics.

In this paper, we develop a quantum smoothing theory to the class of open quantum systems driven by quantum jump-diffusion processes. A quantum smoothing theory, which is derived in a different senses by different researchers, has been applied for diffusive quantum Markovian systems, however, they only deal with quantum Wiener noises. We show how to derive the quantum smoother for quantum dynamical systems driven by quantum jump-diffusion noises.

In this paper, we investigate the design of controllers for the finite-time stabilization of stochastic systems. In the stochastic finite-time stabilization using controllers designed by previous design methods, the input signals often exhibit the chattering-like behavior. This paper presents another method for designing controllers in stochastic finite-time stabilization problems to avoid the chattering-like behavior. Moreover, we discuss the cause of the chattering-like behavior in the finite-time stabilization through numerical results.

In this paper, we study optimal LQ tracking control problems with preview for linear discrete-time hybrid systems. We adopt an approach to optimize system states and mode distributions simultaneously. We consider optimization problems for averaged systems and averaged performance indices throughout mode distributions. On the contrary to the previous work [16, 17], in this paper, we consider to utilize any preview information of reference signals.

In this paper, we study Nash tracking game problems with preview for linear continuous-time systems over finite time intervals. We introduce cross-coupled differential Riccati equations (CDREs) to give solutions of the tracking problems with multi-players. Corresponding to the coupled properties of the CDREs, the forms of preview compensator dynamics are cross-coupled. In this paper, we focus on the derivation of the necessary conditions for the solvability of the tracking problems by variational approach.

In this paper, we study Nash tracking problems for linear continuous-time Markovian jump systems over finite time intervals. We introduce double coupled differential equatuons (CDREs) to give solutions of the tracking problems with multi-players. Corresponding to the double coupled properties of the CDREs, the forms of preview compensator dynamics are double coupled. In this paper, we focus on the derivation of the necessary conditions for the solvability of the tracking problems by variational approach.

In this paper, we are concerned with a problem of optimization of the linear observations which are used in the stationary Kalman-Bucy filter. Especially, we consider the optimization of the gain matrix in the observation. In the previous works the author, by introducing a change of variables using eigenvalues and eigenvectors of a symmetric matrix, the condition of optimality of these variables was obtained. In this paper, it is shown that the expression of the optimal gain obtained in the previous works is unique, and there does not exist any other gain matrix which produces the same value of the cost function. It will be also shown in this paper that under what condition the optimal gain fails to exist and that how one can avoid the situation by modifying the weight matrix in the cost function.

This paper considers an observer-based control for linear systems having outputs quantized by uniform quantizers, where the state-feedback control is applied using the states estimated by the observer. The observer gain is designed to mitigate the effects of the quantization error on the system output. Specifically, we design the optimal observer gain to minimize the H₂ norm of the transfer function from the quantization error to the system output as a function of the observer gain. Numerical examples are provided to see the performance of our design.

The filtering and smoothing techniques for nonlinear continuous-discrete model which is applicable to physical state estimation problems are explained. The multiple distribution estimation and distribution selection are applied to design the non-Gaussian filter. After the execution of the filter, the non-Gaussian smoother is applied to improve the filtered state estimates further. The filtering and smoothing problem for the spacecraft re-entry simulation model is provided as the numerical example and the proposed approach is compared with the other non-Gaussian filters and smoothers.

We proposed a new robust states and parameters estimator for nonlinear systems in this paper. The proposed estimator is derived based on Approximated Unbiased Minimum Variance (AUMV) estimator [3] and Intrinsically Bayesian Robust Kalman filter (IBR-KF) [4]. So, the proposed method takes over the merits of both method. That is, the estimated states are not influenced by the parameter estimation error as in the AUMV estimator, and the estimated parameter is robust against observation noise like the IBR-KF.We confirm the validity of the proposed methods by numerical simulations.

This paper is concerned with the State-of-charge (SOC) estimation for lithium-ion batteries by the Extended Kalman Fixed Lag Smoother (EKFLS). The Extended Kalman Filter (EKF) has been often used as an SOC estimation method for lithium-ion batteries. We employ the EKFLS for the SOC estimation focusing on the fact that the EKFLS can utilize more information than the EKF for the estimation. The effectiveness of the proposed method is verified by a charge-discharge experiment on a 18650 type lithium-ion battery.

One of the effective methods for solar irradiance prediction is based on the Just-In-Time (JIT) modeling technique. To use in micro grid management, this paper presents two techniques to improve the prediction accuracy of the one-day ahead JIT-based prediction method due to Suzuki et al. (2011). The first technique is to update the solar irradiance prediction more frequently in accordance with the updating period of weather forecast data. The second technique is to make a correction to the JIT-based prediction online based on the most recent solar irradiance measurements.

We verify the effectiveness of the proposed techniques by numerical results.

Recently, a distributed cooperative full-state observer for discrete-time linear systems, where it utilizes a consensus mechanism which enables local observers to exchange their estimates in an appropriate interval has been proposed. The behavior of its estimation error has been derived and a necessary and sufficient condition such that the error asymptotically converges to zero has been presented. In this paper, the behavior of the observer systems for a case where stochastically data missing will occur is considered. A simple model by using Bernoulli distribution is introduced to represent random data missing events. Based on the model including a stochastic factor, a necessary and sufficient condition is presented for the mean square stability of the correlation of estimate error. That means its estimation error converges to zero almost surely. Numerical examples are shown to illustrate the behavior of the observer systems where its estimation error converges to zero even if data missing occurs at random.

Various models exist for representing the opinion formation process in social networks. The DeGroot consensus model, which is one of the earliest works in opinion dynamics, updates the opinions of the members of a group by taking into account their interrelationships and how much they value each other's opinions. In this model, the interactions and the weights given to opinions remain fixed during the entire duration of the process, which leads to a predictable behavior. As such, there are certain scenarios that can be better represented using other models. For asynchronous pairwise interactions, a gossip algorithm can be used as an alternative method for describing its opinion dynamics. This approach randomly selects the interacting pairs at every step of the algorithm. Thus, the opinion formation in this model is driven by a stochastic process. We propose to extend the pairwise gossip algorithm to accommodate asynchronous group interactions. We then analyze its behavior and compare it with the other models discussed in this paper.

In this paper, the Schrödinger equation in quantum mechanics is derived as stochastic quantization from the stochastic optimal control theory based on the Nelson stochastic mechanics.

This paper is concerned with the stability analysis of the stochastic age-structured infectious model. We have been faced with a threat of the infectious disease even in the modern society with a high medical technology. Hence, infection prevention and control of epidemic are cited as one of the important social problems. Until now, interesting and impactful research of infectious diseases has been performed using various kinds of mathematical models. However, most of the research is based on the deterministic model and the age-structure has not been taken into account. In the realistic spread of the infectious disease, environmental change and individual difference cause some kinds of random fluctuations in the infection, the recovery rates and the vaccination effect. Moreover, there exists the difference in the transmission intensity of infection by age. Taking these facts into consideration, we propose the stochastic age-structured infectious model. Since the spread of infection has reference to the stability of the disease-free steady state (DFS) of the stochastic infectious models, we analyze the stability of the DFS by using the stochastic Lyapunov theorem.

A stochastic weighted averaging consensus algorithm is considered for a multi-agent system over a noisy unbalanced directed network. The convergence of the algorithm is investigated, which gives an explicit relation between the number of iterations and the closeness of the agreement, i.e., a stopping rule.

This paper deals with a joint input-output method of identifying a closed-loop Hammerstein-Wiener system, where a linear plant in the forward loop may be unstable. A model that simulates the closed-loop system is derived using basis function expansions of nonlinearities. First, a linear state-space model of the forward loop is identified using Two-stage ORT method [12]¹. Then, initialized by the identified state-space model and linear approximations of nonlinearities, the joint input-output mean square error (MSE) is optimized by using the robust gradient method [20]; the identification algorithm yields good numerical results within moderate computation time.

This paper investigates the problem of identifying errors-in-variables (EIV) models, where the both input and output measurements are corrupted by white noises, and addresses a new efficient recursive algorithm. The identification problem of EIV models with unknown noise variances has been extensively studied and several methods have been proposed. To be further developed in terms of estimation accuracy, the bias compensated weighted least squares (BCWLS) method with only requirement of input noise variance estimate has been proposed by using the biased weighted least squares estimate. However, the recursive form for the standard least squares estimate cannot be applied to recursively compute the BCWLS estimate because the weight matrix is not diagonal. To recursively compute the BCWLS estimate, the recursive forms for the biased WLS estimate and the input noise variance estimate are derived. The results of a simulated example indicate that the proposed recursive algorithm provides good results.

We propose two tests for distinguishing nonlinear stochasticity from linear stochasticity and nonlinear determinism based on a time series. One method makes a distinction between stochasticity and determinism using permutations, while the other method highlights the differences between linearity and nonlinearity using surrogate data and test statistics, which are not directly related to determinism or predictability. We analyze the validity of these two tests using toy models.

In this study, we define stochastic models for tight and loose groove feelings and propose an evaluation system for the groove feelings evoked by drum sounds. There are two main types of groove feelings, tight and loose, and they are known among professional musicians and music enthusiasts. These groove models are defined as a bivariate normal distribution with deviations for snare and bass drums as random variables. As an application of these models, we propose our evaluation system, in which a monaural, polyphonic drum set recordings is input and evaluations regarding tight and loose are output. We conducted an evaluation experiment and showed that this system correctly identifies tight and loose groove feelings.

The objective of this paper is to present an algorithm to estimate the relative ranges, angle-of-arrivals (AoAs), and Doppler frequencies of the discrete clutter signals depending on the measurements by a single passive receiver. A single moving emitter source is assumed in this paper. The discrete clutter signals are generated through the multipath reflections of the emitter signal. The technical challenges are identified and an algorithm to resolve them is presented. The high potential of the algorithm is demonstrated by simulation.

A signal processor which is suitable for distilling a melody played by a particular instrument from a music ensemble is discussed in this report. The processor separates a monaural music sound into individual tones of every pitch with a particular timbre, combinations of which provide melodies played by an instrument. A mixture of strongly correlated sounds such as the same pitch tones by different instruments can be processed by the proposed technique. It is a modified Wiener filter for the mixture, the current statistics of which are sequentially evaluated in the chromatic scale with the Gabor wavelet transform.

When problems with strong dependency among decision variables are optimized, a characteristic distribution, which is a thin elliptical distribution, may appear. In order to generate good children, it is necessary to change the variables (genes) simultaneously along the long axis of the elliptical distribution. Since binomial crossover in differential evolution determines whether each gene is crossed or not with the same probability, it is difficult to change some genes simultaneously. In this study, we propose a crossover operation GBX which uses correlation coefficients of search points in order to detect such distribution. The highly correlated genes are grouped and the genes in each group are crossed (or not crossed) simultaneously. However, if only GBX is used, the diversity of the search points tends to be lost rapidly. The adaptive control of the probability for applying GBX is also proposed. The advantage of the proposed method is shown by solving several bench- mark problems.

In this study, we propose a novel multimodal optimization (MMO) algorithm, gravitational particle swarm algorithm (GPSA). This replaces the global feedback term of a classical particle swarm optimization with a term that introduces inverse-square gravitational force between particles. We analyze GPSA's search behavior by Monte-Carlo simulation and evaluate its search performance for a one-dimensional MMO problem. It is shown that our GPSA's particles exhibit gathering and scattering behavior, based on which our GPSA was able to find 83% of the optimal solutions without using any algorithmic clustering procedures.

An algorithm for finding the equilibrium price such that demand and supply coincide is proposed. In the algorithm, the auctioneer publishes a noisy price in order to protect consumers' privacy, while the consumers and the firm bid their optimal demands and the supply at given noisy price. In a conventional work, it is assumed that utility functions and a cost function are quadratic concave and quadratic convex, respectively. In this paper, the assumption is relaxed to strictly concave and strictly convex. Then, it is shown that the algorithm protects consumers' privacy and the candidate price converges to the equilibrium price in a weak sense.

In this paper, the author investigates numerically a class of two-dimensional fuzzy random sets(abbreviated as FRSs), which is one case of those proposed by the author[1] as models of vague capricious perceptions of crisp random phenomena.

First, the basic results concerned with two-dimensional FRSs as vague capricious perceptions of random phenomena are reviewed. Expectations of two-dimensional FRSs, their estimators, and distances between expectations and their estimators are also reviewed briefly.

Finally, estimates for expectations of two-dimensional FRSs and their estimation errors are numerically examined by simulation studies, when their level sets are given by polygon-form ones at every level.

Electrooculographic (EOG) signal reflects the eye movement. This paper proposes the detection of eye movements by EOG signal for a communication of people with movement disabilities. The system consists of a computer, analogue-to-digital converter, digital-to-analogue converter, a direct current (DC) amplifier and a monitor. The monitor is used to display a target for acquiring the eye-gaze position. For adjusting the DC amplifier, the eye-gaze position can be acquired. The target position on the monitor is changed at constant interval, and then the user looks at the target, relation of eye-gaze position and eye potential is derived. From the above, it was verified whether the DC amplifier can be adjusted and horizontal eye movements are detected. Then, the proposed system was used to normal healthy subjects. The result, eye movements is detected, and it can be expected to be used for the development of future communication aid system.

The sweeping robot plans path and moves along its prior path. However, the outdoor sweeping robot cannot move as expected. The uncertain movement of the robot is caused by some environmental disturbance such as unsmooth loose soil and wind. It can be improved by removing the disturbance, however it is difficult to remove the disturbance. For this reason, we consider path planning considering robot uncertainty because of environmental disturbance. This study considers only wind blow as a disturbance. We perform the simulation for sweeping entire area to investigate the effect on the prior path by the wind blow. Simulation results shows it is difficult to turn in the tailwind for the robot, and the moving cost is large. Moreover, it needs to turn for a long time to reach in target position when the angle between the robot direction and the direction to next target position from the robot is large, and the moving cost is large.

In recent years, satellite navigation systems have been actively developed in the world, and they are called GNSS (Global Navigation Satellite System) which includes Global Positioning System (GPS) of the United States, Russian Global Navigation Satellite System (GLONASS), European Galileo (GAL), Chinese BeiDou Navigation Satellite System (BDS) and Japanese Quasi-Zenith Satellite System (QZSS). In this paper, we focus on the so-called multi GNSS precise point positioning (PPP), and propose a positioning method to improve its performance.

According to our previous research works, we present the algorithms of the VPPP (Very Precise Point Positions) using multiple GNSS antennas with known distances each other. VPPP algorithms contain the very precise GNSS positioning as well as the so-called GNSS Gyro by estimating the multiple baseline vectors from the reference antenna to other antennas. In this paper, we attempt to implement the satellite orbits and clocks correction data from the MADOCA Products which had been developed by JAXA, into our VPPP algorithms. Further, finally, we show the positioning accuracy as well as the estimation accuracy of the Euler's angles in the GNSS Gyro, in comparison with using the so-called GPS broadcast ephemeris.

In this paper, for the GNSS (Global Navigation Satellite System) positioning, we propose a method to distinguish non-line-of-sight (NLOS) signals from line-of-sight (LOS) signals by utilizing the satellite geometry and the image data of the fish-eye view of the zenith direction at the antenna position. The NLOS signal is the diffracted signal by buildings or structures, and it greatly degrades the positioning accuracy. By applying the proposed method, we can exclude the NLOS signal and can improve the positioning results in urban areas. In recent years, image recognition using deep learning has developed rapidly. We use the semantic segmentation method using deep learning for segmentation distinguishing sky area from obstacle area.

Real option approaches are handled for reliability evaluations based on degradation models. In general, the reliability analysis is focused on a random variable of life data, and the Weibull distribution is used as a reliability modeling. However a degradation process is significant for making value of the reliability. In particular, stochastic process models are useful for representations of the degradation paths. In this paper, a hierarchical Bayesian VG (Variances Gamma) model is proposed for the representation of the degradation phenomenon, and the real option method based on a simulation is applied to the reliability evaluation using the model. It is also shown that other reliability approaches for the degradation analysis are handles based on GARCH (Generalized Autoregressive Conditional Heteroskedasticity) and SV(Stochastic Volatility) models, and a method of comparison for the models based on the information criterion EIC is proposed.

A novel learning automaton with 2 state Bayesian estimators, which is based on β-type learning automaton, is considered. The β-type learning automaton with multiple Bayesian estimators is our original work and the scheme of it has been proven to be conditionally optimal under some environments. Furthermore it was shown that β-type learning automaton exhibits superior learning behavior compared to conventional learning automata through several experimental results. However, the β-type one is inferior from the viewpoints of memory usage and computational efficiency. So, the β-type learning automaton which consists of 2-state Bayesian estimators as minimum resources has been proposed in our previous work. In this study, we show the sufficient condition for conditional optimality of the β-type LA which consists of 2-state Bayesian estimators.

Currently, we are developing English pronunciation training system for Japanese adults. In this system, pronunciation is indicated by displaying formants, lip shapes, and tongue shapes that are effective for identifying vowels. In this work, focusing on the visualization of tongue shape and lip shape, analysis of MR image and speech signal was performed. Generally, there is a correspondence between the first formant and the opening length of the mouth. Therefore, we obtained correlation coefficients of the first formant detected from speech signal and opening of lips and jaws detected from MR image. As a result, it was found that some cases showed strong correlation or not depending on subjects. From past research, it is known that the Japanese vowel /u/ which is the lingual tongue vowel may change to the front tongue due to the protrusion of the lip. From this, it is considered that the subject who don't show strong correlation manipulates the phoneme by projecting the lips instead of opening the mouth vertically. In order to support this hypothesis, it is necessary to analyze more MR images and speech signals in the future.

The recent aquarium arts simultaneously provide visitors with interest in science technology and fish ecosystems. In particular, some studies on soft robotics develops some fish robots with a feeling similar to reality. These robots have the potential to be a new life that can coexist with human and actual fish. This study aims to develop a soft fish robot that allows people to directly contact and control swimming easily. We introduce 3D printed mold, magnetic soft fish mixed with silicone rubberizing reagent and sand iron. The concept of real fish swarm using magnetic field is proposed on this paper. In this paper, we show a basic follow-up of fish robot and verify soft fish swarm by moving magnetic field. The results reveal that the fish robot follows the magnet due to the magnetic change of the sand iron inside the silicon. Refer to the principle of boids, the fish swarm shows that the dynamic mutual interference between fishes by magnetic sand iron can be described as a cohesion of magnetic force, separation of collision and alignment by a water flow.