2023 IEEE/SICE International Symposium on System Integration (SII)最新文献

{"title":"Hardware-in-the-loop Simulation for Real-time Autonomous Tracking and Landing of an Unmanned Aerial Vehicle","authors":"Nobuaki Aoki, G. Ishigami","doi":"10.1109/SII55687.2023.10039438","DOIUrl":"https://doi.org/10.1109/SII55687.2023.10039438","url":null,"abstract":"Collaborative exploration using unmanned aerial vehicles (UAVs) and unmanned ground vehicles (UGVs) has become increasingly popular in the past decade. This study addresses real-time autonomous tracking and landing of a UAV on a moving ground vehicle, which is fundamental for the UAV-UGV collaborative exploration. The method proposed in this study estimates the relative pose and velocity between a UAV and UGV, and uses model predictive control for UAV trajectory planning while considering the field of view of the camera onboard the UAV. We elaborate the hardware-in-the-loop simulator (HITL) with a physical companion computer, and confirm that the proposed method enables a UAV to land on a UGV traversing on rough terrain based on online computations in the HITL. Additionally, we present statistical analysis of the simulation results of typical and computationally demanding scenarios to elucidate the computational cost on the real machine.","PeriodicalId":421076,"journal":{"name":"2023 IEEE/SICE International Symposium on System Integration (SII)","volume":"46 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-01-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114798958","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":"A Deep-Learning Approach to Marble-Burying Quantification: Image Segmentation of Marbles and Bedding","authors":"Yicheng Zhu, Brandon Hudson, Chandranil Chakraborttii, Yun-Hsuan Su, Kevin Huang","doi":"10.1109/SII55687.2023.10039320","DOIUrl":"https://doi.org/10.1109/SII55687.2023.10039320","url":null,"abstract":"This paper presents and evaluates three automated tools for semantically segmenting images from marble-burying experiments. The marble-burying animal model is widely used as an indication of anxiety or obsessive compulsive behavior in rodents. In general, the tendency for caged rodents to bury objects in their bedding is seen as anxiety related, and several methods have been proposed to measure the degree of this burying behavior. Unfortunately, most of these methods are coarse or require either subjective interpretation or onerous manual procedures. Digital imaging can provide pre- and post-experiment burying states as well as a platform for a standardized and streamlined quantification of the marble-burying test. While continuous imaging streams might provide more information and temporal analysis, such datasets are rare, require expert annotation, and can be prohibitively large. The authors propose that single-image semantic, pixel-wise segmentation of marble and bedding pixels are key components that can enable effective quantification. For example, the ratio of marble to bedding pixels can provide greater granularity in assessing marble-burying behavior. In this work, a classical image segmentation approach, a single-class U-Net and a multi-class U-Net were comparatively evaluated via standard segmentation metrics. Results show that the deep-learning methods demonstrate greater segmentation performance than the traditional method. Timing-performance trade off considerations between single- and multi-class methods are also explored.","PeriodicalId":421076,"journal":{"name":"2023 IEEE/SICE International Symposium on System Integration (SII)","volume":"26 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-01-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134394793","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":"Classification of pathological and healthy individuals for computer-aided physical rehabilitation","authors":"Ivanna Kramer, Raphael Memmesheimer, D. Paulus","doi":"10.1109/SII55687.2023.10039185","DOIUrl":"https://doi.org/10.1109/SII55687.2023.10039185","url":null,"abstract":"The role of computer-aided therapy in physical rehabilitation has significantly grown in recent years. In order to optimize the therapeutic actions based on the patient disease and improve the interaction between the patient and telemedical software, it is important to differentiate the healthy individuals and patients following different chronic diseases or musculoskeletal disorders. In this paper, we propose a deep learning method to classify the trajectory patterns in physical rehabilitation as healthy or pathological. Motion sequences containing joint positions and joint angles are transformed into an image representation, which enables the training of a classification model using a deep 2D Convolutional Neural Network (CNN) to infer a health state or a symptom. Our approach was evaluated on two publicly available datasets, the KIMORE [1] dataset for complete body skeleton sequences and the Toronto Rehab Stroke Posture (TRSP) [2] dataset containing upper body skeleton sequences. The current method effectively classifies the healthy and pathological movements and achieves 78.57% of accuracy on KIMORE dataset and 86.20% of accuracy on TRSP data.","PeriodicalId":421076,"journal":{"name":"2023 IEEE/SICE International Symposium on System Integration (SII)","volume":"27 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-01-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133082547","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":"Reinforcement Learning-based Motion Generation for a Tracked Robot to Go Over a Sphere-shaped Non-fixed Obstacle","authors":"Hidenori Takamiya, Ryosuke Yajima, Junichi Kasahara, Ren Komatsu, K. Nagatani, A. Yamashita, H. Asama","doi":"10.1109/SII55687.2023.10039098","DOIUrl":"https://doi.org/10.1109/SII55687.2023.10039098","url":null,"abstract":"Tracked robots have high traversability over rough terrain. However, even for such robots, it is still challenging to traverse terrain with non-fixed obstacles which may move when the robots go over them. Therefore, we propose a reinforcement learning-based method to generate the motion of the tracked robot to go over the obstacle. We set a task where the robot attempts to go over a sphere-shaped non-fixed obstacle and reach the goal. To succeed in the task, we designed a reward function so that the robot can reach the goal as straight as possible. As a training algorithm, Deep Q-Network was used and the robot was trained in a dynamics simulator. It was confirmed that the robot succeeded in the task using the trained network, which generated motion for going over a sphere-shaped non-fixed obstacle.","PeriodicalId":421076,"journal":{"name":"2023 IEEE/SICE International Symposium on System Integration (SII)","volume":"5 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-01-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115422730","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":"Fabrication of a Spherical-Shape Dielectric Elastomer Actuator Made of an Organogel","authors":"Y. Tezuka, Asahi Tsujino, T. Hayakawa","doi":"10.1109/SII55687.2023.10039186","DOIUrl":"https://doi.org/10.1109/SII55687.2023.10039186","url":null,"abstract":"In this study, we fabricated three-dimensional (3D) DEA made of an organogel and drove it out-of-plane direction using an experimental system that applied pre-stretch with water pressure. Organogels have a lower Young's modulus and higher relative permittivity than elastomer materials used in conventional DEA such as PDMS. Therefore, fabrication of DEAs using organogel is expected to increase the displacement of DEAs. Thus, we fabricated a dielectric gel that is a kind of organogel with high dielectric constant and excellent elasticity. We evaluated static characteristics and frequency response of water-pre-stretched DEA made of dielectric gel. We also fabricated three-dimensional dielectric gel for a 3D shaped DEA. From a driving experiment, it was confirmed that this 3D DEA can generate 10 times larger displacement than that of PDMS.","PeriodicalId":421076,"journal":{"name":"2023 IEEE/SICE International Symposium on System Integration (SII)","volume":"12 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-01-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130173822","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":"Camera-based Progress Estimation of Assembly Work Using Deep Metric Learning","authors":"Takumi Kitsukawa, Sarthak Pathak, Alessandro Moro, Yoshihiro Harada, Hideo Nishikawa, Minori Noguchi, A. Hamaya, K. Umeda","doi":"10.1109/SII55687.2023.10039109","DOIUrl":"https://doi.org/10.1109/SII55687.2023.10039109","url":null,"abstract":"In this paper, a progress estimation method using deep learning is proposed to visualize the product assembly process in a factory. First, the target assembly product is detected from images acquired from a fixed-point camera installed in the factory using a deep learning-based object detection method. Next, the detection area is cropped from the image. Finally, by using a classification method based on deep metric learning on the cropped image, the progress of the product assembly work is estimated as a rough progress step. In addition, considering the similarity of features with neighboring steps when learning with deep metric learning, we propose an adaptive loss function that learns to separate features from nearby steps. In experiments, an 82 [%] success rate is achieved for the progress estimation method using deep metric learning. Furthermore, the method using the adaptive loss function achieved a success rate of 92 [%]. Experiments were also conducted to verify the practicality of a series of detection, cropping and progress estimation.","PeriodicalId":421076,"journal":{"name":"2023 IEEE/SICE International Symposium on System Integration (SII)","volume":"125 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-01-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132394007","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":"Attitude and Orbit Control Strategy with 10N Thruster for ALE-3","authors":"Kazuki Kibune, S. Fujita, Y. Shibuya, Toshinori Kuwhara, Hiromune Ishii, Pasith Tangdhanakanond","doi":"10.1109/SII55687.2023.10039171","DOIUrl":"https://doi.org/10.1109/SII55687.2023.10039171","url":null,"abstract":"This paper presents a low earth orbit control sequence for a microsatellite using thruster control and its evaluation. The Space Robotics Laboratory (SRL) and ALE Co.,Ltd. are currently developing the ALE-3 with the mission of artificial generation of a meteor. Since the meteor ejection mission will be performed at an altitude of about 400 km or lower, ALE-3 will perform a descent from an altitude of 600 km using 10N thrusters. Due to the limited amount of propellant loaded onto the satellite, efficient thruster operation is important to achieve the target orbit. In this study, we propose a control sequence that achieves highly efficient orbit transfer by using other actuators for attitude control before and after thruster injection. Simulation results show that orbit control using the proposed method can change the satellite's velocity with high efficiency and the propellant mounted on the satellite is sufficient to achieve the target orbit transition.","PeriodicalId":421076,"journal":{"name":"2023 IEEE/SICE International Symposium on System Integration (SII)","volume":"52 7","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-01-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132899087","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":"Study on Portable Respiratory Assist System for Independence Living of Neuromuscular Intractable Disease Patients","authors":"Tatsuaki Nemoto, H. Kawamoto, Akira Uehara, Y. Sankai","doi":"10.1109/SII55687.2023.10039328","DOIUrl":"https://doi.org/10.1109/SII55687.2023.10039328","url":null,"abstract":"In neuromuscular intractable diseases, respiratory dysfunction due to muscle atrophy requires assistance from a large, heavy ventilator. Even though patients with respiratory dysfunction retain the ability to walk, they are forced to use a wheelchair, resulting in a vicious cycle of decreased independence and muscle group disuse. Additionally, ventilation using an improperly timed ventilator can cause pain and discomfort to the user. The daily use of a ventilator requires a device that is non-invasive and portable and respiratory assistance that is synchronized with the user's respiratory timing. The purpose of this study is to develop an assistive method for voluntary respiration that is synchronized with the user's spontaneous respiration on a daily basis using a system that is small, lightweight, portable, and non-invasive in measuring respiratory intention and supporting respiration. The noninvasive voluntary respiratory assist system consists of a control unit, air sensor module, centrifugal microblower, sensor unit for bio-electrical signals (BES), and removable battery. The weight is approximately 2 kg, which is easy to carry. For the noninvasive detection of respiratory intention, we developed three algorithms: (1) for detecting spontaneous respiration using flow rate values in a nasal mask, (2) for detecting mouthpiece holding movements, and (3) for detecting spontaneous respiration using BES of the external intercostal muscles. The results of a basic experiment with one participant showed that at a pressure of 20 cmH2O the system is able to support a higher tidal volume than that of an adult. The spontaneous respiration detection algorithm, using the flow rate value in the nasal mask, and the mouthpiece holding motion detection algorithm enabled voluntary respiratory assistance without error in the resting sitting and walking conditions. Furthermore, the spontaneous respiration detection algorithm using the respiratory muscle BES enabled voluntary respiratory assistance without error in the resting sitting condition. These results indicate that the system can provide voluntary respiratory assistance that is synchronized with the user's spontaneous respiration on a daily basis.","PeriodicalId":421076,"journal":{"name":"2023 IEEE/SICE International Symposium on System Integration (SII)","volume":"85 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-01-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122275621","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":"Shielded Reinforcement Learning: A review of reactive methods for safe learning","authors":"Haritz Odriozola-Olalde, Maider Zamalloa, N. Arana-Arexolaleiba","doi":"10.1109/SII55687.2023.10039301","DOIUrl":"https://doi.org/10.1109/SII55687.2023.10039301","url":null,"abstract":"Reinforcement Learning (RL) algorithms are showing promising results in simulated environments, but their replication in real physical applications, even more so in safety-critical applications, is not yet guaranteed. Ensuring the functional safety of RL algorithms is not a trivial task since the physical integrity of the target system, also called environment, especially when there is interaction with humans, may depend on it. Among the methods recently developed with the objective of guaranteeing safety, Shielded Reinforcement Learning is defined, which defines an interaction mechanism if the action event proposed by the agent causes a non-safe state. This article provides an overview of the different Shielding Reinforcement Learning approaches. In addition to summarising the techniques used by each of them, their advantages and disadvantages are discussed. Finally, the shortcomings associated with Shielded Reinforcement Learning methods that can lead to risk or unsafe situations are discussed.","PeriodicalId":421076,"journal":{"name":"2023 IEEE/SICE International Symposium on System Integration (SII)","volume":"54 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-01-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121725838","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":"Near-infrared iPPG body motion compensation using head Euler angles during sleep","authors":"Keisuke Terai, T. Ohhira, Hideki Hashimoto","doi":"10.1109/SII55687.2023.10039132","DOIUrl":"https://doi.org/10.1109/SII55687.2023.10039132","url":null,"abstract":"Image photoplethysmography (iPPG) is a non-invasive and non-contact method that can measure the heart rate (HR) of a person while sleeping without causing discomfort. However, the iPPG signal can be contaminated by body motion, making HR measurement difficult. Therefore, body motion noise caused by tossing and turning during sleep can cause sections where HR cannot be measured. Thus, we propose a method of recovering the iPPG signal contaminated by body motion using the x and y coordinates of the region of interest in the face and the Euler angle of the head. The proposed system reduces the influence of tossing and turning during sleep and improves the accuracy of HR measurement. In our experiment, subjects were instructed to sleep at stationary and mixed body movement positions (turning every 15 s) in an environment designed for sleeping. The experimental results confirmed that the HR measurement accuracy was improved over conventional methods both at rest and when body motions occurred.","PeriodicalId":421076,"journal":{"name":"2023 IEEE/SICE International Symposium on System Integration (SII)","volume":"79 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-01-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133456653","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}