研究者詳細 - 迎田　隆幸

Various man-machine interfaces controlled by electromyogram (EMG) signals such as the myoelectric prosthetic hand have been proposed. General classifiers can not consider unintended motions in the training phase and require learning all the motions. In this paper, the authors propose a motion estimation system with unlearned classes and combined motions based on a muscle synergy model. The proposed method can identify unlearned five-finger combined motions by learning a single motion only. Furthermore, this method utilizes the history of muscle synergy and unlearned motion detection, using a state transition model. In the experiments, it was shown that the discrimination accuracy was sufficient for simple combined motions.

ロバストな作業識別のための余事象動作分類と隠れセミマルコフモデルを用いた介護作業識別

島田悠之介, 迎田隆幸, 日下聖, 遠藤維, 多田充徳, 宮田なつき, 田中孝之

計測自動制御学会論文集 60 ( 12 ) 620 - 630 2024年

DOI CiNii Research

記述言語：日本語掲載種別：研究論文（学術雑誌）出版者・発行元：公益社団法人計測自動制御学会共著

In this paper, we propose a work description by elemental motion and a work identification method for creating detailed care records. This method describes tasks as a time-series of elemental motion and identifies care works. In order to achieve robust work identification against irregular motion and annotation of incorrect motion labels during a work, some elemental motion are integrated into one class as complementary event motion. Using the time-series data of elemental motion information, our proposed work identification method utilizing Hidden Semi-Markov Models can recognize performed care works based on the frequency distribution of state transitions. In the experiment, three simulated care works were measured and identified. In addition, simulations using artificial data were conducted to verify the robustness of the proposed method. Furthermore, we measured actual caregiving tasks at a nursing home to confirm whether the task detection was feasible. The results demonstrated the effectiveness of the proposed system.

FPGA実装を指向した未学習クラス推定混合ガウス型識別モデルと複合動作の識別

柏木僚太, 迎田隆幸, 島圭介

計測自動制御学会論文集 60 ( 6 ) 397 - 406 2024年

DOI CiNii Research

記述言語：日本語掲載種別：研究論文（学術雑誌）出版者・発行元：公益社団法人計測自動制御学会共著

The electromyogram (EMG) signal generated by muscle contraction has been widely utilized for motion estimation of arms and fingers. To develop a myoelectric prosthetic hand that has high general versatility and safeness, a classifier that can consider complex forearm motions and motions that are not assumed during training, is required. However, hardware implementation of complex classifiers that has high classification performance is difficult. To tackle these problems, this paper proposed a novel probabilistic neural network that can be implemented in FPGA (Field Programmable Gate Array), and it was applied to an EMG-based human-machine interface system. The proposed neural network includes two types of probability density functions optimized for hardware implementation and enabled the execution of multi-class discrimination considering the unlearned class on the FPGA. Furthermore, by combining a forearm motion classifier and a hand motion classifier, the consideration of compound motions consisting of multiple hand gestures can be achieved. In experiments, the results showed that the proposed method can be implemented on FPGAs, and demonstrated that it can achieve highly accurate motion classification for compound motions and unlearned motions.

FPGA Implementation of Approximate Gaussian Mixture Model for Open-Set Recognition in Interface Control

Kashiwagi, R; Mukaeda, T; Shima, K

2024 10TH INTERNATIONAL CONFERENCE ON CONTROL, DECISION AND INFORMATION TECHNOLOGIES, CODIT 2024 1553 - 1558 2024年

DOI Web of Science

記述言語：日本語掲載種別：研究論文（学術雑誌）共著

Open-set motion recognition and adaptive structural modification of classifiers based on clustering of unknown motions

Takayuki Mukaeda, Keisuke Shima

2023 IEEE International Conference on Systems, Man, and Cybernetics 2023年10月 [査読有り]

DOI

担当区分：筆頭著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）単著

乳牛ルーメンモデルの開発と隠れセミマルコフモデルを用いた行動モデリング

石川優理矢, 田中孝之, 迎田隆幸, 松田朝陽, 石川志保

ロボティクス・メカトロニクス講演会講演概要集 2023 ( 0 ) 2A1-A26 2023年

DOI CiNii Research

記述言語：日本語掲載種別：研究論文（学術雑誌）出版者・発行元：一般社団法人日本機械学会共著

In dairy farming, it is required to raise animals in consideration of animal welfare. In addition, it is also required to increase the amount of milk, which is the product. For these realizations, it is necessary for humans to intervene theoretically and cows to be relaxed. Therefore, we defined an index of satiety from the stomach contents of dairy cows. In addition, a behavioral model was created using a hidden semi-Markov model. By introducing an index of satiety and creating a behavior model, it was confirmed that the behavior model changed appropriately for each satiety degree. When hungry, it transitioned to eating behavior, and when full, it transitioned to non-eating behavior. We were able to create a model that matches the actual behavior.

ウェアラブル・エンベデッドセンサを用いた作業姿勢分類と作業検知への応用

迎田隆幸, 島田悠之介, 田中孝之, 野口宏明, 阿部敏久

計測自動制御学会論文集 58 ( 12 ) 558 - 567 2022年

DOI CiNii Research

記述言語：日本語掲載種別：研究論文（学術雑誌）出版者・発行元：公益社団法人計測自動制御学会共著

To evaluate the work contents of a care worker, this paper proposes a novel lower limb posture estimation method, and a motion recognition method for the automation of care records. In our approach, wearable-embedded sensors consisting of inertial sensors and insole-type plantar pressure sensors were used to recognize the posture of the whole body during care tasks. The upper body posture was calculated from triaxial accelerations, and posture classification of lower limbs can be reached with characteristics extracted from plantar pressure distribution. To achieve accurate posture recognition considering unexpected posture in training phase and compound postures combining multiple postures, a new posture estimation method combining normal and complementary Gaussian mixture network (NACGMN) and posture-fitness function was developed. Using the time-series data of posture information, our proposed work estimation manner utilizing hidden semi-Markov models can recognize performed care works based on the frequency distribution of state transitions. In experiments, we measured and recognized transfer assistance works between a bed and a wheelchair in our laboratory and an actual care facility, and the results demonstrated the effectiveness of the proposed system.

混合余事象分布を内包した確率ニューラルネットに基づく5指駆動型ロボットハンドのEMG制御

小宮山翼, 迎田隆幸, 島圭介

計測自動制御学会論文集 57 ( 12 ) 504 - 510 2021年

DOI CiNii Research

記述言語：日本語掲載種別：研究論文（学術雑誌）出版者・発行元：公益社団法人計測自動制御学会共著

Many man-machine interfaces controlled by electromyogram (EMG) signals such as the myoelectric prosthetic hand have been proposed. General classifiers do not cover unintended motions in the training phase and misclassify those inevitably. Since the misclassification can cause dangerous incidents, an interface with high security is required. To solve this problem, this paper proposes a novel control method of man-machine interfaces that can treat unlearned motions. In the experiments, the proposed method was applied to forearm and finger motion classification to evaluate the validity. The outcomes showed that the approach produces higher and more stable classification performance than comparative methods.

溶接作業熟練者のクラスタ構造解析と作業評価システムの開発

迎田隆幸, 島圭介, 池田裕樹, 遠藤直輝, 野々村将一, 藤井正和

ロボティクス・メカトロニクス講演会講演概要集 2021 ( 0 ) 2P3-J16 2021年

DOI CiNii Research

記述言語：日本語掲載種別：研究論文（学術雑誌）出版者・発行元：一般社団法人日本機械学会共著

In recent years, the shortage of successors has become a serious problem in the manufacturing industry due to the declining birthrate and the aging of skilled workers. To tackle this problem, a new education and training systems utilizing ICT technology that improves the efficiency of education is required. In this paper, we focused on the education of welding work and proposed a new work evaluation system that visualizes arc welding operations and evaluates the similarity with skilled workers. The proposed method can extract low-dimensional feature vectors from measured welding torch postures and current and voltage values, and accurately classifies them into multiple clusters acquired by unsupervised learning. The experimental results showed that the proposed system achieved high classification performance and suggested the possibility of using them for training in welding operations.

A System for Wound Evaluation Support Using Depth and Image Sensors

Watanabe Ryotaro, Shima Keisuke, Horiuchi Taiki, Shimizu Takeshi, Mukaeda Takayuki, Shimatani Koji

2021 43RD ANNUAL INTERNATIONAL CONFERENCE OF THE IEEE ENGINEERING IN MEDICINE & BIOLOGY SOCIETY (EMBC) 3709 - 3712 2021年

DOI Web of Science PubMed

記述言語：日本語掲載種別：研究論文（学術雑誌）共著

混合余事象分布に基づく未学習クラス推定確率ニューラルネット

迎田隆幸, 島圭介

計測自動制御学会論文集 56 ( 12 ) 532 - 540 2020年

DOI CiNii Research

記述言語：日本語掲載種別：研究論文（学術雑誌）出版者・発行元：公益社団法人計測自動制御学会共著

General classification methods only involve consideration of learned classes, and do not cover undefined targets such as unintended characteristics in the learning process. This paper proposes a novel probabilistic neural network can treat unlearned class. The proposed method incorporates two types of probabilistic distribution: normal and complementary Gaussian distribution and can reach multi-class classification and unlearned class detection with a single network. In the experiments, artificial data and electromyogram (EMG) patterns were classified to demonstrate the capabilities of the proposed method. The results showed that the approach produces high performance for classification, and there were significant differences between the proposed and previous methods.

Real-time evaluation of driver cognitive loads based on multivariate biosignal analysis

Shimizu Takeshi, Shima Keisuke, Mukaeda Takayuki, Muraji Shu, Matsuo Juntaro, Horiue Masayoshi

2020 IEEE INTERNATIONAL CONFERENCE ON SYSTEMS, MAN, AND CYBERNETICS (SMC) 623 - 628 2020年

DOI Web of Science

記述言語：日本語掲載種別：研究論文（学術雑誌）共著

FPGA実装を指向した近似GMMに基づく未学習クラス推定ニューラルネット

植草秀明, 迎田隆幸, 清水武史, 島圭介

ロボティクス・メカトロニクス講演会講演概要集 2020 ( 0 ) 1P2-F05 2020年

DOI CiNii Research

記述言語：日本語掲載種別：研究論文（学術雑誌）出版者・発行元：一般社団法人日本機械学会共著

General classification methods cannot treat unexpected patterns that are not considered in the training process. For this problem, our research group has proposed a probabilistic classification method that can classify abnormal patterns as the unlearned class. For the implementation of the classifier in the embedded hardware, this paper proposes a novel approximate Bayesian classification method with the anomaly detection based on Gaussian mixture models and the probabilistic density function of the unlearned class. The proposed classifier was applied to forearm motion classification in the experiments. Experimental results demonstrate the proposed method can achieve high classification performance as same as the previous model, and the effectiveness of the proposed method could be confirmed.

Development of an anomaly detection method with a novel hidden semi-Markov model incorporating unlearned states

Mukaeda Takayuki, Shima Keisuke, Miyajima Saori, Hashimoto Yuki, Tanaka Takayuki, Tani Naomichi, Iz … 全著者表示

2020 IEEE/SICE INTERNATIONAL SYMPOSIUM ON SYSTEM INTEGRATION (SII) 1270 - 1275 2020年

DOI Web of Science

記述言語：日本語掲載種別：研究論文（学術雑誌）共著

Effects of somatosensory information provision to fingertips for mitigation of postural sway and promotion of muscle coactivation in an upright posture

Mitani Ryoma, Shimatani Koji, Sakata Mami, Mukaeda Takayuki, Shima Keisuke

2019 41ST ANNUAL INTERNATIONAL CONFERENCE OF THE IEEE ENGINEERING IN MEDICINE AND BIOLOGY SOCIETY (EMBC) 5096 - 5099 2019年

Web of Science

記述言語：日本語掲載種別：研究論文（学術雑誌）共著

隠れマルコフモデルに基づく未学習クラスの推定法と時系列生体信号の識別

迎田隆幸, 島圭介

計測自動制御学会論文集 54 ( 1 ) 9 - 15 2018年 [査読有り]

DOI

担当区分：筆頭著者記述言語：日本語掲載種別：研究論文（学術雑誌）単著

A novel hidden Markov model-based pattern discrimination method with the anomaly detection for EMG signals

T. Mukaeda and K. Shima

2017 39th Annual International Conference of the IEEE Engineering in Medicine and Biology Society 921 - 924 2017年 [査読有り]

DOI

担当区分：筆頭著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）単著

このページの先頭へ▲

<迎田 隆幸>

論文 - 迎田 隆幸

<迎田　隆幸>

論文 - 迎田　隆幸