Publishing type：Research paper (international conference proceedings)  

The Volterra filter is a digital filter that can describe nonlinearity. In this work, we analyze the dynamic behaviors of an adaptive signal processing system with the Volterra filter and a time-varying unknown system by a statistical-mechanical method. Specifically, assuming the self-averaging property with an infinitely long tapped-delay line, we derive simultaneous dif-ferential equations that describe the behaviors of the macroscopic variables in a deterministic and closed form and obtain the exact solution by solving them analytically. In addition, the validity of the derived theory is confirmed by comparison with numerical simulations.

Scopus

researchmap

Authorship：Corresponding author   Publishing type：Research paper (scientific journal)   Publisher：The Society of Instrument and Control Engineers  

DOI： 10.9746/sicetr.56.31

researchmap

Publishing type：Research paper (scientific journal)  

© 2019 The Author(s) Nagata et al. proposed a parameter estimation method using Markov chain Monte Carlo (MCMC) for the spectral deconvolution of observed data. However, a systematic error occurs when the parameters to be estimated are close. In this paper, we first clarify that the exchange symmetry of parameters, which is essentially included in the spectral deconvolution problem, causes the systematic error. In particular, we show that estimation from a single data set is inherently difficult because the posterior distribution becomes unimodal or multimodal depending on the data set when the parameters to be estimated are close. Second, we alleviate the problem to the case of using multiple data sets and propose an extension of the exchange Monte Carlo method to low temperatures. This extension corresponds to bridging the gap between posterior mean (PM) estimation and maximum a posteriori (MAP) estimation. The above alleviation and bridging achieve a good estimation even when the parameters are close.

DOI： 10.1016/j.neunet.2019.05.004

Scopus

PubMed

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：Institute of Electrical and Electronics Engineers Inc.  

The Volterra filter is a digital filter that can describe nonlinearity. In this paper, we analyze the dynamic behaviors of an adaptive signal-processing system based on the Volterra filter by a statistical-mechanical method. On the basis of the self-averaging when the tapped delay line is assumed to be infinitely long, we derive simultaneous differential equations in a deterministic and closed form that describe the behaviors of the macroscopic variables and obtain the exact solution by solving them analytically. In addition, the validity of the derived theory is confirmed by comparison with numerical simulations.

DOI： 10.23919/APSIPA.2018.8659649

Scopus

researchmap

Publishing type：Research paper (scientific journal)   Publisher：Japan Society of Mechanical Engineers  

DOI： 10.1299/jsmermd.2019.2a2-c05

researchmap

Authorship：Lead author   Language：Japanese   Publishing type：Research paper (scientific journal)   Publisher：The Institute of Systems, Control and Information Engineers  

DOI： 10.5687/iscie.31.202

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：Institute of Physics Publishing  

A mobile robot with two independently driven wheels is popular, but it is difficult to stabilize it by a continuous controller with a constant gain, due to its nonholonomic property. It is guaranteed that a nonholonomic controlled object can always be converged to an arbitrary point using a switching control method or a quasi-continuous control method based on an invariant manifold in a chained form. From this, the authors already proposed a kinodynamic controller to converge the states of such a two-wheeled mobile robot to the arbitrary target position while avoiding obstacles, by combining the control based on the invariant manifold and the harmonic potential field (HPF). On the other hand, it was confirmed in the previous research that there is a case that the robot cannot avoid the obstacle because there is no enough space to converge the current state to the target state. In this paper, we propose a method that divides the final target position into some sub-Target positions and moves the robot step by step, and it is confirmed by the simulation that the robot can converge to the target position while avoiding obstacles using the proposed method.

DOI： 10.1088/1742-6596/962/1/012012

Scopus

researchmap

Language：Japanese   Publishing type：Research paper (scientific journal)   Publisher：Institute of Electrical Engineers of Japan (IEE Japan)  

DOI： 10.1541/ieejeiss.138.369

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：IEEE  

The autonomous control of the quadrotor has received attention. In the previous research, we proposed the kinodynamic control using a harmonic potential field (HPF), which is a method to design a control input by considering kinematic constraints and dynamic constraints simultaneously. However, it was assumed that the environment was already known and the HPF of the environment was prepared in advance. Therefore, in this paper, assume that a final target position is given to the quadrotor and the quadrotor moves by relying on the mounted sensor. It is confirmed, by a simulation using the Matlab, that the proposed method can guide the quadrotor to the target position.

Web of Science

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：IEEE  

We analyze the behaviors of active noise control using a statistical-mechanical method. The principal assumption used in the analysis is that the impulse responses of the primary path and adaptive filter are sufficiently long. In particular, in this paper we analyze the adaptation rate of the mean square error (MSE) using two measures. The first measure is the MSE initial decreasing rate. The second measure is an adaptation constant. This is defined by the negative of the maximum eigenvalue of the coefficient matrix of differential equations that describe the dynamical behaviors of the macroscopic variables. Introducing these two measures, we theoretically show that the optimal step size depends on whether we focus on the rate of decrease in the MSE at the initial stage or the MSE after sufficient adaptation time.

Web of Science

researchmap

Authorship：Lead author   Language：Japanese   Publishing type：Research paper (scientific journal)   Publisher：Japan Society of Mechanical Engineers  

DOI： 10.1299/transjsme.14-00631

researchmap

Language：English   Publishing type：Part of collection (book)   Publisher：IGI Global  

This chapter describes kinodynamic motion planning and its application. Kinodynamics is the discipline that tries to solve kinematic constraints and dynamical constraints simultaneously. By using kinodynamic motion planning, control inputs can be generated in a much simpler way, compared to the conventional motion planning that solves kinematics and dynamics separately. After briefly overviewing the kinodynamic motion planning, its application to a flying robot is described in detail.

DOI： 10.4018/978-1-4666-7387-8.ch015

Scopus

researchmap

Publishing type：Research paper (scientific journal)  

A method of automatically optimizing gains is described for kinodynamic motion planning related to a controlled system consisting of a point mass. Kinodynamic motion planning proposed by Masoud has some gains and it is difficult to optimize such gains manually due to its interaction. Note, however, that any method for optimizing the gains has not been mentioned yet. Therefore, a method for optimizing all gains included in the kinodynamic motion planning is proposed by using a genetic algorithm. © 2013 ISAROB.

DOI： 10.1007/s10015-013-0129-6

Scopus

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：IEEE  

In this research, it is aimed at guiding an X4-Flyer, which is a VTOL type UAV, to an arbitrary target point. In the previous research, we proposed a controller for an X4-Flyer to move on the X-Y plane by using a 2-dimensional harmonic potential field (HPF), assuming that the X4-Flyer keeps its altitude constantly. In this paper, the controller guides the X4-Flyer to an arbitrary target point in a 3-dimensional space by switching some 2-dimensional HPFs appropriately. It is confirmed on the simulation that the X4-Flyer can reach the arbitrary target point in the 3-dimensional space by using the proposed method.

DOI： 10.1109/ICCAS.2014.6987738

Web of Science

researchmap

Publishing type：Research paper (scientific journal)  

Under the assumption that sensors are distributed in the environmental side to measure human movement in a term, such as a couple of days, we first propose a human frequency map (HFM), which is a grid map based on the observed human position and frequency in the term. Then, using such an HFM and the distributed sensor data, the possibility of encountering human and the width of passage in addition to the current position of human are taken into account in our new path planning method. The usefulness of the proposed path planning approach is demonstrated through simulations in dynamic environments, as well as actual experiments to show the realization of safe robot navigation. © 2013 Sociedade Brasileira de Automatica - SBA.

DOI： 10.1007/s40313-013-0011-8

Scopus

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：Society of Instrument and Control Engineers (SICE)  

In this research, we present a method to control an X4-Flyer using kinodynamic motion planning, and optimize each gain for the controller to obtain more stable control to a target point. In this paper, for realizing kinodynamic motion planning, we propose a controller combined nonholonomic control, which is used for an existing controller for X4-Flyer, with harmonic potential field (HPF). It is difficult to optimize gains manually, because there are a number of gains in the controller. Therefore, each gain is optimized by using GA. Then, it is confirmed that the X4-Flyer can also be controlled in an environment different from the searched environment by using searched gains.

Scopus

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：ALIFE ROBOTICS CO, LTD  

A robot requires any path planning to respect human movements when the robot works with the person in an environment. Therefore, we use a Human Frequency Map (HFM) generated by using observed human positions. By applying the HFM, the robot can achieve the path planning that cares about the human movements and the pathway width. This paper describes the effectiveness of the HFM through some experiments in the real environment. The experimental results of navigation using a real mobile robot is also discussed.

Web of Science

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：SOC INSTRUMENT CONTROL ENGINEERS JAPAN  

A mobile robot used in home or office has to not only search the shortest path but also perform the path planning to respect human movement. In this study, several sensors are distributed in the environmental side, and humans walking route is observed. Then, we propose a Human Frequency Map (HFM) that is grid map based on the observed human position. By using the HFM, the proposed path planning can minimize the possibility for encountering human by considering the human frequency of a certain time zone and the current position of human. In this paper, we evaluate the usefulness of the information obtained from the HFM for the path planning in the case where the robot follows the specified person while avoiding the other persons.

Web of Science

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)  

A path planning is studied for a mobile robot that works in a living environment. If the robot moves in the living environment, then the robot has to achieve the path planning so as to care about the persons as well as recognize the environment. Assume here that sensors are distributed in the environmental side, and human's walking is observed. The human's walking route is safe and easier to move for the robot because the human avoids obstacles or selects the efficient route. Thus, in this research, observed human positions are used for the path planning of the robot. Then, a "Human Frequency Map (HFM)" is proposed, where it is a grid map based on the observed human position and frequency in a term. In addition to the current position of human, the possibility of encountering human and the width of passage can be considered to perform a path planning by using the HFM.

Scopus

researchmap

Event date： 2016.9

Venue：Korea  

researchmap

Event date： 2016.1

Venue：Oita, Japan  

researchmap

Event date： 2015.12

Venue：名古屋  

researchmap

Event date： 2015.11

Venue：Yokohama, Japan  

researchmap

Event date： 2015.10

Venue：Makassar, Indonesia  

researchmap

Event date： 2015.9

Venue：東京  

researchmap

Event date： 2014.12

Venue：Fukuoka, Japan  

researchmap

Event date： 2014.11

researchmap

Event date： 2014.10

researchmap

Event date： 2014.9

Venue：Makassar, Indonesia  

researchmap

Event date： 2014.9

Venue：Hokkaido, Japan  

researchmap

Event date： 2014.9

researchmap

Event date： 2014.5

Venue：富山  

researchmap

Event date： 2013.12

Venue：兵庫  

researchmap

Event date： 2013.11

Venue：Tokyo, Japan  

researchmap

Event date： 2013.11

Venue：Vienna, Austria  

researchmap

Event date： 2013.9

researchmap

Event date： 2013.9

Venue：東京  

researchmap

Event date： 2013.8

Venue：Kuantan, Malaysia  

researchmap

Event date： 2013.1

Venue：Daejeon, Korea  

researchmap

Event date： 2012.8

Venue：Akita, Japan  

researchmap

Event date： 2012.5

Venue：静岡  

researchmap

Event date： 2012.1

Venue：Oita, Japan  

researchmap

Event date： 2011.12

Venue：Phuket, Thailand  

researchmap

Event date： 2010.12

Venue：宮城  

researchmap

Event date： 2010.12

Venue：Okayama, Japan  

researchmap

researchmap

researchmap

researchmap

researchmap

Venue：Shenzhen, China  

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap