{"title":"Control of HVAC-Systems Using Reinforcement Learning With Hysteresis and Tolerance Control","authors":"C. Blad, C. Kallesøe, Simon Boegh","doi":"10.1109/SII46433.2020.9026189","DOIUrl":"https://doi.org/10.1109/SII46433.2020.9026189","url":null,"abstract":"This paper presents the idea of using tolerance control in Deep Reinforcement Learning to improve robustness and reduce training time. This paper is a continuation of [1] where it is shown that Reinforcement Learning (RL) can be used to control an underfloor heating (UFH) system. However, it is seen in the study that the initial training time is too high and that the performance during training is not fulfilling the requirements to a UFH system. In this paper the fundamental challenge regarding control of UFH systems is explained, how RL can be beneficial for control of UFH systems, and how the implementation is done. Furthermore, results are presented with a standard hysteresis control, an RL control, and an RL control with tolerance control. These results show that the effect of tolerance control in these types of systems is significant. Finally, we discuss the challenges there are for a real-world implementation of RL-based control in UFH system.","PeriodicalId":253265,"journal":{"name":"2020 IEEE/SICE International Symposium on System Integration (SII)","volume":"432 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115599988","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Analysis of sound activities and voice activity detection using in-car microphone arrays","authors":"C. Ishi, A. Utsumi, Isamu Nagasawa","doi":"10.1109/SII46433.2020.9026235","DOIUrl":"https://doi.org/10.1109/SII46433.2020.9026235","url":null,"abstract":"In this study, we evaluate the collaboration of multiple microphone arrays installed in the interior of a car, with the aim of robustly identifying the driver’s voice activities embedded in car environment noises. We first conducted preliminary analysis on the identified sound activities from the sound direction estimations by different microphone arrays arranged under the physical constraints of the car interior. Driving audio data was collected under several car environment conditions, including engine noise, road noise, air conditioner, winker sounds, radio sounds, driver’s voice, passenger voices, and external noises from other cars. The driver’s voice activity intervals could be identified with 97% detection rate by combining two microphone arrays, one around the “eyesight” camera system cover and the other around the driver’s sun visor.","PeriodicalId":253265,"journal":{"name":"2020 IEEE/SICE International Symposium on System Integration (SII)","volume":"19 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115706963","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Hiroki Dobashi, Masaya Fukunishi, Masaya Minami, Y. Yokokohji
{"title":"Minimization of Actuator’s Driving Force of Chuck-type Hand with Three Parallel Stick Fingers through Grasp Optimization","authors":"Hiroki Dobashi, Masaya Fukunishi, Masaya Minami, Y. Yokokohji","doi":"10.1109/SII46433.2020.9025839","DOIUrl":"https://doi.org/10.1109/SII46433.2020.9025839","url":null,"abstract":"For realizing a versatile, flexible robotic assembly system in manufacturing, development of a versatile robotic hand with minimally necessary mechanism has been expected. As a simple hand for this demand, we are currently aiming at developing a chuck-type hand with three parallel stick fingers which can achieve a target assembly task. Toward the optimal design of the hand, it is necessary to know the required driving force of an actuator to be used. This paper presents a method to compute the magnitude of the minimally necessary driving force of the actuator for the hand through grasp optimization.","PeriodicalId":253265,"journal":{"name":"2020 IEEE/SICE International Symposium on System Integration (SII)","volume":"20 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125178774","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Yasuto Yokota, Kanata Suzuki, Yuzi Kanazawa, T. Takebayashi
{"title":"A Multi-task Learning Framework for Grasping-Position Detection and Few-Shot Classification","authors":"Yasuto Yokota, Kanata Suzuki, Yuzi Kanazawa, T. Takebayashi","doi":"10.1109/SII46433.2020.9025946","DOIUrl":"https://doi.org/10.1109/SII46433.2020.9025946","url":null,"abstract":"It is a big problem that a model of deep learning for a picking robot needs many labeled images. Operating costs of retraining a model becomes very expensive because the object shape of a product or a part often is changed in a factory. It is important to reduce the amount of labeled images required to train a model for a picking robot. In this study, we propose a multi-task learning framework for few-shot classification using feature vectors from an intermediate layer of a model that detects grasping positions. In the field of manufacturing, multitask for shape classification and grasping-position detection is often required for picking robots. Prior multi-task learning studies include methods to learn one task with feature vectors from a deep neural network (DNN) learned for another task. However, the DNN that was used to detect grasping positions has two problems with respect to extracting feature vectors from a layer for shape classification: (1) Because each layer of the grasping position detection DNN is activated by all objects in the input image, it is necessary to refine the features for each grasping position. (2) It is necessary to select a layer to extract the features suitable for shape classification. To tackle these issues, we propose a method to refine the features for each grasping position and to select features from the optimal layer of the DNN. We then evaluated the shape classification accuracy using these features from the grasping positions. Our results confirm that our proposed framework can classify object shapes even when the input image includes multiple objects and the number of images available for training is small.","PeriodicalId":253265,"journal":{"name":"2020 IEEE/SICE International Symposium on System Integration (SII)","volume":"79 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126184315","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Z. Zhong, Katsutoshi Itoyama, Kenji Nishida, K. Nakadai
{"title":"Design and Assessment of a Scan-and-sum Beamformer for Surface Sound Source Separation","authors":"Z. Zhong, Katsutoshi Itoyama, Kenji Nishida, K. Nakadai","doi":"10.1109/SII46433.2020.9025981","DOIUrl":"https://doi.org/10.1109/SII46433.2020.9025981","url":null,"abstract":"This paper proposes a novel robot audition technique for listening to a surface sound source. A beamforming technique called scan-and-sum beamformer is proposed. It performs sound source separation of general wide band surface sources distributed in the azimuth angle dimension. Because a sound source is mainly modeled as a point source in current signal processing technology, it is difficult for a robot or system with a conventional beamformer to address a surface source, though most sound sources in the real world are surface sources with a certain shape and size. The basic idea of scan-and-sum beamforming is to decompose a surface source into numerous point sources intensively distributed within a region, and then integrate results from point source sub-beamformers. In the scan-and-sum approach, a point source sub-beamformer scans with appropriate scanning density a region where target surface sources exist, and sub-beamformers are summed to separate the surface sound sources. Due to the high flexibility of the proposed beamformer, a mean square error (MSE) criterion for error analysis and cost function J to balance performance and cost are introduced, with guidelines on how to implement a scan-and-sum beamformer with tuned parameters and optimized pattern. Compared to conventional point source beamformers, beam patterns with a broader mainlobe and lower sidelobes can be achieved at acceptable cost. Numerical simulations show the proposed method improves signal-to-interference ratio (SIR) for a mixture of three surface sound sources.","PeriodicalId":253265,"journal":{"name":"2020 IEEE/SICE International Symposium on System Integration (SII)","volume":"45 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123767103","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Akira Uehara, H. Kawamoto, H. Imai, M. Shirai, M. Sone, Sachiko Noda, Shigeto Sato, N. Hattori, Y. Sankai
{"title":"Lateral Swing Support System for Parkinsonism patients with Freezing of Gait","authors":"Akira Uehara, H. Kawamoto, H. Imai, M. Shirai, M. Sone, Sachiko Noda, Shigeto Sato, N. Hattori, Y. Sankai","doi":"10.1109/SII46433.2020.9025882","DOIUrl":"https://doi.org/10.1109/SII46433.2020.9025882","url":null,"abstract":"Freezing of gait (FOG) is one of the typical locomotion disorders associated with Parkinsonism, and is described as the inability to move the feet despite an attempt to walk. The fear of falling over and other symptoms of FOG can be distressing to patients. Considering conventional external cues and physical therapy, it is necessary for Parkinsonism patients suffering from FOG to swing the body laterally and rhythmically. The purpose of this study is to develop a wearable system that can assist rhythmical and lateral swing, and to validate the performance of the system in preventing FOG by conducting a gait experiment on a Parkinsonism patient. To develop this system, we simulated lateral swing in the standing position based on features of Parkinsonism and defined a mechanism and appropriate force configurations for the system. A power unit at the back of the waist transmits a cyclic force to the chest through a link in the frontal plane. A gait experiment was performed on a Parkinsonism patient with FOG to confirm the reduction in duration time and number of steps of turning. The result showed that the cyclic assistance of the system and gait cycle were harmonized, and the developed system decreased duration time and number of steps during turning by half. In addition, Pre-FOG, i.e., transient small stepping gait, occurred when turning without using the system, while it did not occur during and immediately after the use of the system. These results suggest that the developed system is effective in preventing occurrences of FOG by providing lateral and cyclic force to the upper body.","PeriodicalId":253265,"journal":{"name":"2020 IEEE/SICE International Symposium on System Integration (SII)","volume":"129 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115187153","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"IntelliBike: An Intelligent Bicycle with Automatic Shifting Based on Sensing and Riding Analysis","authors":"Zong-Yao Wu, H. Lin","doi":"10.1109/SII46433.2020.9026293","DOIUrl":"https://doi.org/10.1109/SII46433.2020.9026293","url":null,"abstract":"This paper presents an IntelliBike with automatic shifting based on the riding condition. It is able to collect and transmit multiple sensor data to the cloud database, and the real-time sensor information is used to analyze for an appropriate riding state through random forest. Our technique performs automatic gear shifting using inexpensive sensor components and embedded system development board, which is not available in current expensive electronic shifting systems. The function and the installation of the external components can be carried out on the off-the-shelf bicycle with a shifting line, without the need to dismantle or modify the basic equipment. The system also includes an IntelliBike-specific online management interface that allows all cloud data to be displayed and let users access and understand their riding conditions.","PeriodicalId":253265,"journal":{"name":"2020 IEEE/SICE International Symposium on System Integration (SII)","volume":"13 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116146563","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Eiichirou Tanaka, Jyun-Rong Zhuang, Bo-Rong Yang, G. Wu, Hee-Hyol Lee, L. Yuge
{"title":"A Walking Assistive Device of the Ankle Joint Motion and the Control Method According to the Emotion Condition","authors":"Eiichirou Tanaka, Jyun-Rong Zhuang, Bo-Rong Yang, G. Wu, Hee-Hyol Lee, L. Yuge","doi":"10.1109/SII46433.2020.9026187","DOIUrl":"https://doi.org/10.1109/SII46433.2020.9026187","url":null,"abstract":"A walking assistive device which induces only ankle joint motion has been developed. This device is attached on the ankle joint and only assisting the motion of it at the adequate phase, whole leg can be raised spontaneously. It was already sold as a product “RE-Gait®” for hemiplegic patients walking training. Now the inexpensive version of both leg for the elderly is also developed, which can make users to improve their gait. Furthermore, we propose to control the device taken into account the emotion condition of the user to promote and continue walking. We developed the current emotion evaluation system expressed with two axes; arousal and pleasant, using physiological data. On the other hand, the gait also can be expressed with two axes; cadence and step length. By integrating this system and device, we confirmed the relation between the gait and emotion, and it can assist the users walking effectively.","PeriodicalId":253265,"journal":{"name":"2020 IEEE/SICE International Symposium on System Integration (SII)","volume":"28 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116574977","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"3D Hole Center and Surface Normal Estimation in Robot Vision Systems","authors":"Ja-Won Seo, Kyung-Eun Kim, Kyungshik Noh","doi":"10.1109/SII46433.2020.9026272","DOIUrl":"https://doi.org/10.1109/SII46433.2020.9026272","url":null,"abstract":"In this paper, we present a novel stereo camera-based 3D hole center and surface normal estimation method which can compute the relative pose (i.e., position and orientation) of a hole to a camera precisely. Unlike conventional methods which are completely dependent on either the geometrical model of a target or the visual feedback, the proposed method takes a novel open-loop approach on the basis of specific stereo image processing. Therefore, the computational speed of the proposed method is high enough to be applicable to real-time tasks in industrial applications while maintaining requirements for the localization accuracy. These breakthroughs are achieved by suppressing 3D geometrical errors from the stereo image processing based on the well-known DLT (Direct Linear Transformation) technique. In extensive experiments, the proposed method shows very promising performance in terms of both processing speed and pose estimation accuracy in a typical robot vision system which is equipped with a binocular eye-in-hand apparatus.","PeriodicalId":253265,"journal":{"name":"2020 IEEE/SICE International Symposium on System Integration (SII)","volume":"85 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122047338","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Frequency analysis of heart rate variability with equalized power spectrum for each heartbeat","authors":"Koji Yoshida, K. Wada","doi":"10.1109/SII46433.2020.9026286","DOIUrl":"https://doi.org/10.1109/SII46433.2020.9026286","url":null,"abstract":"The method for evaluating the human psychological state is employed in several fields. This evaluation generally involves the state of functioning of the autonomic nerve. Monitoring this state continuously and characterizing changes of the autonomic nerve facilitates recognition of changes of the human psychological state. Quantitative comparison of each frequency component of the autonomic nervous system per unit time is necessary to monitor these changes. Also, the RR interval, which is the difference in the R-wave generation time in ECG, is a fluctuation in the time axis direction that is, the x axis direction. Therefore, it is difficult to directly analyze the frequency of the observed data. In general, amplitude data is assessed against time before frequency analysis is performed. In this study, the amplitude is defined by considering the sum of the power spectral density of each beat is equal. Comparing the representation method by the conventional method revealed differences in the frequency component ratio. This indicates that some errors are method dependent.","PeriodicalId":253265,"journal":{"name":"2020 IEEE/SICE International Symposium on System Integration (SII)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129707433","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}