Pub Date : 2022-12-05DOI: 10.1109/ROBIO55434.2022.10011959
Yujian Zhang, Yuan Liu, Shiyin Qiu, Fengrui Ji, Jinze Wei, Dong Ming
Motor imagery-based brain-computer interfaces (MI-BCI) help patients to reconstruct damaged neural path-ways in the field of neurorehabilitation. However, difficulties in performing abstract imagery tasks and generating discriminable EEG signals for some subjects limit the application of MI-BCI, and the devices required for the visual guidance paradigm are not portable in MI-BCI application scenarios for wearable robotic systems. In this study, we propose an enhanced motor imagery paradigm combining sequential elec-trical stimulation (SES) encoded by gait phase with a gait motor imagery (MI) task, guiding subjects to perform MI task with task-mapped electrical stimulation (ES). The goal of the novel paradigm is to reduce the difficulty of lower limbs MI task and to improve the performance of the MI-BCI by combining movement and sensation. We conducted comparison experiments on eight healthy subjects, and the MI task in the SES-Stim paradigm achieved greater activation of motor cortex in the $alpha$ and $beta$ rhythm, and the proposed SES-Stim paradigm could improve the classification performance.
{"title":"Enhanced Motor Imagery of Lower Limbs Induced by Gait Phase Encoding Sensory Electrical Stimulation","authors":"Yujian Zhang, Yuan Liu, Shiyin Qiu, Fengrui Ji, Jinze Wei, Dong Ming","doi":"10.1109/ROBIO55434.2022.10011959","DOIUrl":"https://doi.org/10.1109/ROBIO55434.2022.10011959","url":null,"abstract":"Motor imagery-based brain-computer interfaces (MI-BCI) help patients to reconstruct damaged neural path-ways in the field of neurorehabilitation. However, difficulties in performing abstract imagery tasks and generating discriminable EEG signals for some subjects limit the application of MI-BCI, and the devices required for the visual guidance paradigm are not portable in MI-BCI application scenarios for wearable robotic systems. In this study, we propose an enhanced motor imagery paradigm combining sequential elec-trical stimulation (SES) encoded by gait phase with a gait motor imagery (MI) task, guiding subjects to perform MI task with task-mapped electrical stimulation (ES). The goal of the novel paradigm is to reduce the difficulty of lower limbs MI task and to improve the performance of the MI-BCI by combining movement and sensation. We conducted comparison experiments on eight healthy subjects, and the MI task in the SES-Stim paradigm achieved greater activation of motor cortex in the $alpha$ and $beta$ rhythm, and the proposed SES-Stim paradigm could improve the classification performance.","PeriodicalId":151112,"journal":{"name":"2022 IEEE International Conference on Robotics and Biomimetics (ROBIO)","volume":"46 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125524575","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}
Pub Date : 2022-12-05DOI: 10.1109/ROBIO55434.2022.10011817
Tanhao Zhang, Luyin Hu, Yuxiang Sun, Lu Li, D. Navarro-Alarcon
Compared with 2D thermal images, visualizing the temperature of objects with their corresponding 3D surfaces provides a more intuitive way to perceive the environment. In this paper, we present an integrated system for large-scale and real-time 3D thermographic reconstruction through fusion of visible, infrared and depth images. The system is composed of an RGB-D and a thermal camera, whose image measurements are aligned with respect to the same coordinate frame. A thermal direct method based on infrared features is proposed and integrated into state-of-art localization algorithms for generating reliable 3D thermal point clouds. The reported experimental results demonstrate that our approach can be used for 3D reconstruction of small and large scale environments based on dual spectrum 3D information.
{"title":"Computing Thermal Point Clouds by Fusing RGB-D and Infrared Images: From Dense Object Reconstruction to Environment Mapping","authors":"Tanhao Zhang, Luyin Hu, Yuxiang Sun, Lu Li, D. Navarro-Alarcon","doi":"10.1109/ROBIO55434.2022.10011817","DOIUrl":"https://doi.org/10.1109/ROBIO55434.2022.10011817","url":null,"abstract":"Compared with 2D thermal images, visualizing the temperature of objects with their corresponding 3D surfaces provides a more intuitive way to perceive the environment. In this paper, we present an integrated system for large-scale and real-time 3D thermographic reconstruction through fusion of visible, infrared and depth images. The system is composed of an RGB-D and a thermal camera, whose image measurements are aligned with respect to the same coordinate frame. A thermal direct method based on infrared features is proposed and integrated into state-of-art localization algorithms for generating reliable 3D thermal point clouds. The reported experimental results demonstrate that our approach can be used for 3D reconstruction of small and large scale environments based on dual spectrum 3D information.","PeriodicalId":151112,"journal":{"name":"2022 IEEE International Conference on Robotics and Biomimetics (ROBIO)","volume":"86 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126191029","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}
A novel continuous perception garment classification mechanism is proposed in this paper, with the aim to identify the correct category of the garment from a set of known categories. It has been observed that due to the severe folding and overlapped texture of garments, treating a video of the continuous deformation of cloth as a set of disordered static figures would be ineffective which leads to low classification precision performed by an image-based garment classifier. In contrast, a high-level decision making module that leverages the classification results of each single image would significantly increase the algorithm performance. In this paper, we incorporate the optical flow variation of deformable cloth between consecutive configurations as a representative of how it is traversing within the confidence interval of the image-based classifier. We claim that it is not the number of video frames but the sum of optical flow variation of the garment configuration between consecutive frames having the same category label that constitutes the belief of garment classification. In other words, if two consecutive visual appearances of the garment could be identified as the same category by the image-based classifier, then the more diverged that two configurations are, the more confident that the garment is correctly identified. Experimental comparisons between the state-of-the-art algorithm and the proposed algorithm in a public dataset have been provided which prove the validity of the proposed algorithm.
{"title":"Continuous Perception Garment Classification Based on Optical Flow Variation","authors":"Li Huang, Tong Yang, Yu Zhang, Rongxin Jiang, Xiang Tian, Yao-wu Chen","doi":"10.1109/ROBIO55434.2022.10011783","DOIUrl":"https://doi.org/10.1109/ROBIO55434.2022.10011783","url":null,"abstract":"A novel continuous perception garment classification mechanism is proposed in this paper, with the aim to identify the correct category of the garment from a set of known categories. It has been observed that due to the severe folding and overlapped texture of garments, treating a video of the continuous deformation of cloth as a set of disordered static figures would be ineffective which leads to low classification precision performed by an image-based garment classifier. In contrast, a high-level decision making module that leverages the classification results of each single image would significantly increase the algorithm performance. In this paper, we incorporate the optical flow variation of deformable cloth between consecutive configurations as a representative of how it is traversing within the confidence interval of the image-based classifier. We claim that it is not the number of video frames but the sum of optical flow variation of the garment configuration between consecutive frames having the same category label that constitutes the belief of garment classification. In other words, if two consecutive visual appearances of the garment could be identified as the same category by the image-based classifier, then the more diverged that two configurations are, the more confident that the garment is correctly identified. Experimental comparisons between the state-of-the-art algorithm and the proposed algorithm in a public dataset have been provided which prove the validity of the proposed algorithm.","PeriodicalId":151112,"journal":{"name":"2022 IEEE International Conference on Robotics and Biomimetics (ROBIO)","volume":"41 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125671928","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}
Pub Date : 2022-12-05DOI: 10.1109/ROBIO55434.2022.10011860
Dong Liu, Hongmin Wu, Kezheng Sun, Y. Guan
Anomalous diagnosis is valuable for reducing potential damages in long-term autonomy robot manipulation tasks, especially in Human-robot collaboration scenarios. Deep learning-based methods have been widely investigated for robot anomaly diagnosis, which can effectively encode complex dynamics from multi-modal sensory data. However, the lacking of enough anomalous samples and the fusion of high-dimensional and modality correlation as well as time-dependent is still a challenging problem. In this paper, a novel framework is introduced to generate synthetic anomaly samples for data augmentation by learning the disentangled representation with sequential disentangled variational autoencoder (sDVAE), and a temporal-correlation VAE (tcVAE) model for robot anomaly diagnosis by learning the temporal correlation features of multimodal anomalies. To evaluate the proposed methods, 115 original anomalous samples from 7 representative anomalies that are first recorded on a self-developed human-robot kitting task. Results indicate that the proposed methods show the best performance of the highest precision (97%), f1-score (95%), and accuracy (93%) with synthetic samples across all baseline methods.
{"title":"Learning Disentangled Representations and Temporal-Correlation Dynamics for Robotic Anomaly Diagnosis","authors":"Dong Liu, Hongmin Wu, Kezheng Sun, Y. Guan","doi":"10.1109/ROBIO55434.2022.10011860","DOIUrl":"https://doi.org/10.1109/ROBIO55434.2022.10011860","url":null,"abstract":"Anomalous diagnosis is valuable for reducing potential damages in long-term autonomy robot manipulation tasks, especially in Human-robot collaboration scenarios. Deep learning-based methods have been widely investigated for robot anomaly diagnosis, which can effectively encode complex dynamics from multi-modal sensory data. However, the lacking of enough anomalous samples and the fusion of high-dimensional and modality correlation as well as time-dependent is still a challenging problem. In this paper, a novel framework is introduced to generate synthetic anomaly samples for data augmentation by learning the disentangled representation with sequential disentangled variational autoencoder (sDVAE), and a temporal-correlation VAE (tcVAE) model for robot anomaly diagnosis by learning the temporal correlation features of multimodal anomalies. To evaluate the proposed methods, 115 original anomalous samples from 7 representative anomalies that are first recorded on a self-developed human-robot kitting task. Results indicate that the proposed methods show the best performance of the highest precision (97%), f1-score (95%), and accuracy (93%) with synthetic samples across all baseline methods.","PeriodicalId":151112,"journal":{"name":"2022 IEEE International Conference on Robotics and Biomimetics (ROBIO)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125833339","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}
Pub Date : 2022-12-05DOI: 10.1109/ROBIO55434.2022.10012018
Yameng Zhang, Long Bai, Li Liu, Hongliang Ren, Max Q.-H. Meng
Due to its non-invasive and painless characteristics, wireless capsule endoscopy has become the new gold standard for assessing gastrointestinal disorders. Omissions, however, could occur throughout the examination since controlling capsule endoscope can be challenging. In this work, we control the magnetic capsule endoscope for the coverage scanning task in the stomach based on reinforcement learning so that the capsule can comprehensively scan every corner of the stomach. We apply a well-made virtual platform named VR-Caps to simulate the process of stomach coverage scanning with a capsule endoscope model. We utilize and compare two deep reinforcement learning algorithms, the Proximal Policy Optimization (PPO) and Soft Actor-Critic (SAC) algorithms, to train the permanent magnetic agent, which actuates the capsule endoscope directly via magnetic fields and then optimizes the scanning efficiency of stomach coverage. We analyze the pros and cons of the two algorithms with different hyperparameters and achieve a coverage rate of 98.04% of the stomach area within 150.37 seconds.
{"title":"Deep Reinforcement Learning-Based Control for Stomach Coverage Scanning of Wireless Capsule Endoscopy","authors":"Yameng Zhang, Long Bai, Li Liu, Hongliang Ren, Max Q.-H. Meng","doi":"10.1109/ROBIO55434.2022.10012018","DOIUrl":"https://doi.org/10.1109/ROBIO55434.2022.10012018","url":null,"abstract":"Due to its non-invasive and painless characteristics, wireless capsule endoscopy has become the new gold standard for assessing gastrointestinal disorders. Omissions, however, could occur throughout the examination since controlling capsule endoscope can be challenging. In this work, we control the magnetic capsule endoscope for the coverage scanning task in the stomach based on reinforcement learning so that the capsule can comprehensively scan every corner of the stomach. We apply a well-made virtual platform named VR-Caps to simulate the process of stomach coverage scanning with a capsule endoscope model. We utilize and compare two deep reinforcement learning algorithms, the Proximal Policy Optimization (PPO) and Soft Actor-Critic (SAC) algorithms, to train the permanent magnetic agent, which actuates the capsule endoscope directly via magnetic fields and then optimizes the scanning efficiency of stomach coverage. We analyze the pros and cons of the two algorithms with different hyperparameters and achieve a coverage rate of 98.04% of the stomach area within 150.37 seconds.","PeriodicalId":151112,"journal":{"name":"2022 IEEE International Conference on Robotics and Biomimetics (ROBIO)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129786119","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}
Pub Date : 2022-12-05DOI: 10.1109/ROBIO55434.2022.10011923
Jinzhou Wang, Ran Huang
The ability of mobile robots to navigate in an unfamiliar environment in human terms is decisive for their applicability to practical activities. Bearing this view in mind, we propose a novel framework for navigation in settings where the environment is a priori unknown and can only be partially observed by the robot with onboard sensors. The proposed hierarchical navigation solution combines deep reinforcement learning-based perception with model-based control. Specifically, a deep reinforcement learning (DRL) network based on Soft Actor-Critic (SAC) algorithm and Long Short-Term Memory (LSTM) is trained to map the robot's states, 2D lidar inputs and goal position to a series of local waypoints which are optimal in the sense of collision avoidance. The waypoints are then employed by a dynamic window approach (DWA) based planner to generate a smooth and dynamically feasible trajectory that is tracked by using feedback control. The experiments performed on an actual wheeled robot demonstrate that the proposed scheme enables the robot to reach goal locations more reliably and efficiently in unstructured environments in comparison with purely learning based approach.
{"title":"A Mapless Navigation Method Based on Deep Reinforcement Learning and Path Planning","authors":"Jinzhou Wang, Ran Huang","doi":"10.1109/ROBIO55434.2022.10011923","DOIUrl":"https://doi.org/10.1109/ROBIO55434.2022.10011923","url":null,"abstract":"The ability of mobile robots to navigate in an unfamiliar environment in human terms is decisive for their applicability to practical activities. Bearing this view in mind, we propose a novel framework for navigation in settings where the environment is a priori unknown and can only be partially observed by the robot with onboard sensors. The proposed hierarchical navigation solution combines deep reinforcement learning-based perception with model-based control. Specifically, a deep reinforcement learning (DRL) network based on Soft Actor-Critic (SAC) algorithm and Long Short-Term Memory (LSTM) is trained to map the robot's states, 2D lidar inputs and goal position to a series of local waypoints which are optimal in the sense of collision avoidance. The waypoints are then employed by a dynamic window approach (DWA) based planner to generate a smooth and dynamically feasible trajectory that is tracked by using feedback control. The experiments performed on an actual wheeled robot demonstrate that the proposed scheme enables the robot to reach goal locations more reliably and efficiently in unstructured environments in comparison with purely learning based approach.","PeriodicalId":151112,"journal":{"name":"2022 IEEE International Conference on Robotics and Biomimetics (ROBIO)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129806035","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}
Pub Date : 2022-12-05DOI: 10.1109/ROBIO55434.2022.10011641
Qiqi Li, Xiangrong Xu, Hao Yang, Xiaoyi Wang, Zhixiong Wang, Haiyan Wang, Shanshan Xu, A. Rodic, P. Petrovic
When the manipulator performs the operation task, there are modeling errors and the influence of external disturbance, which is easy to lead to the large tracking error of the manipulator end trajectory. Firstly, according to the structure of the manipulator, the dynamic model of the manipulator is established. Then RBF neural network and self -adaptation are introduced. Compared with the traditional error function, the sliding mode function is introduced in the algorithm, which can ensure the system to approach the desired trajectory quickly. The neural network used has the ability to estimate the uncertainty of the system and reduce the bad influence of interference on the system. Adaptive law and robust term are also introduced to improve the performance of the system. Finally, Lyapunov function is used to prove the stability of the system, and MATLAB/SIMULINK simulation software is used to carry out simulation experiments. Simulation results show that the algorithm has a good effect on disturbance suppression, and the end tracking accuracy is also improved.
{"title":"An Adaptive Control of Manipulator Based on RBF Neural Network Approximation","authors":"Qiqi Li, Xiangrong Xu, Hao Yang, Xiaoyi Wang, Zhixiong Wang, Haiyan Wang, Shanshan Xu, A. Rodic, P. Petrovic","doi":"10.1109/ROBIO55434.2022.10011641","DOIUrl":"https://doi.org/10.1109/ROBIO55434.2022.10011641","url":null,"abstract":"When the manipulator performs the operation task, there are modeling errors and the influence of external disturbance, which is easy to lead to the large tracking error of the manipulator end trajectory. Firstly, according to the structure of the manipulator, the dynamic model of the manipulator is established. Then RBF neural network and self -adaptation are introduced. Compared with the traditional error function, the sliding mode function is introduced in the algorithm, which can ensure the system to approach the desired trajectory quickly. The neural network used has the ability to estimate the uncertainty of the system and reduce the bad influence of interference on the system. Adaptive law and robust term are also introduced to improve the performance of the system. Finally, Lyapunov function is used to prove the stability of the system, and MATLAB/SIMULINK simulation software is used to carry out simulation experiments. Simulation results show that the algorithm has a good effect on disturbance suppression, and the end tracking accuracy is also improved.","PeriodicalId":151112,"journal":{"name":"2022 IEEE International Conference on Robotics and Biomimetics (ROBIO)","volume":"316 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129947630","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}
Pub Date : 2022-12-05DOI: 10.1109/ROBIO55434.2022.10012003
Sihui Liu, Pu Duan, Jianda Han, Ningbo Yu
Hand exoskeletons are attracting rising research interest for assistance in daily life. Friction is a critical issue for hand exoskeletons intended to enable dexterous finger movements. This paper presents a rigid-soft hybrid assistive hand exoskeleton using a low friction mechanism based on the multi-layered springs with rolling contact. The low friction designs not only facilitate the high transparency of the device, but also the possibility of accurate modeling in mechanics. Regarding the large deformation nature of the nonlinear mechanism, finite element analysis (FEA) has been used to explore its kinematic and kinetic characteristics. Experiments were carried out on the implemented prototype to compare the numerical results with the measured. The results demonstrate that this modeling method can provide valuable guidance for better parameter selection and optimization.
{"title":"Design and Evaluation of a Low Friction Rigid-Soft Hybrid Mechanism for Hand Exoskeletons with Finite Element Analysis","authors":"Sihui Liu, Pu Duan, Jianda Han, Ningbo Yu","doi":"10.1109/ROBIO55434.2022.10012003","DOIUrl":"https://doi.org/10.1109/ROBIO55434.2022.10012003","url":null,"abstract":"Hand exoskeletons are attracting rising research interest for assistance in daily life. Friction is a critical issue for hand exoskeletons intended to enable dexterous finger movements. This paper presents a rigid-soft hybrid assistive hand exoskeleton using a low friction mechanism based on the multi-layered springs with rolling contact. The low friction designs not only facilitate the high transparency of the device, but also the possibility of accurate modeling in mechanics. Regarding the large deformation nature of the nonlinear mechanism, finite element analysis (FEA) has been used to explore its kinematic and kinetic characteristics. Experiments were carried out on the implemented prototype to compare the numerical results with the measured. The results demonstrate that this modeling method can provide valuable guidance for better parameter selection and optimization.","PeriodicalId":151112,"journal":{"name":"2022 IEEE International Conference on Robotics and Biomimetics (ROBIO)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129453224","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}
Pub Date : 2022-12-05DOI: 10.1109/ROBIO55434.2022.10011909
Kexin Xiang, Weiqun Wang, Z. Hou, Chutian Zhang, Jiaxing Wang, Weiguo Shi, Yuze Jiao, Tianyu Lin
Accurate rehabilitation assessments are essential for designing effective rehabilitation methods and helping patients recover better. It's well known that commonly used scale assessment methods for neurorehabilitation suffer from the issue of subjectivity, thus investigation of objective assessment methods is very necessary. Muscle synergy analysis can be uesd to assess limb motor functions from the perspective of neuromuscular control. In this paper, a method for evaluation of human lower limb motor functions based on muscle synergy analysis is presented. Muscle synergy modules are designed using surface electromyography (sEMG) signals of the subjects' lower limbs by non-negative matrix factorizations (NMF). By comparing the cosine similarities of these synergy modules, it can be seen that muscle synergies of healthy subjects and patients are significantly different, while they are similar among healthy subjects. Therefore, a reference synergy module (RSM) is designed by averaging the muscle synergy modules for healthy subjects, and the similarities can be calculated by comparing the synergy modules for healthy subjects or patients with the RSM. In the experiment carried out in this study, average similarities of the three synergy modules for healthy subjects are respectively 0.97166, 0.87368 and 0.84932, and on the other hand, the average similarities for the three synergy modules for patients are respectively 0.59979, 0.56426 and 0.69042. Therefore, the similarities for healthy subjects are much higher than those for SCI patients, which denotes that the similarity between an individual synergy module and the RSM can be used as an objective assessing index for evaluating patients' motor function.
{"title":"Muscle Synergy Analysis Based on NMF for Lower Limb Motor Function Assessment","authors":"Kexin Xiang, Weiqun Wang, Z. Hou, Chutian Zhang, Jiaxing Wang, Weiguo Shi, Yuze Jiao, Tianyu Lin","doi":"10.1109/ROBIO55434.2022.10011909","DOIUrl":"https://doi.org/10.1109/ROBIO55434.2022.10011909","url":null,"abstract":"Accurate rehabilitation assessments are essential for designing effective rehabilitation methods and helping patients recover better. It's well known that commonly used scale assessment methods for neurorehabilitation suffer from the issue of subjectivity, thus investigation of objective assessment methods is very necessary. Muscle synergy analysis can be uesd to assess limb motor functions from the perspective of neuromuscular control. In this paper, a method for evaluation of human lower limb motor functions based on muscle synergy analysis is presented. Muscle synergy modules are designed using surface electromyography (sEMG) signals of the subjects' lower limbs by non-negative matrix factorizations (NMF). By comparing the cosine similarities of these synergy modules, it can be seen that muscle synergies of healthy subjects and patients are significantly different, while they are similar among healthy subjects. Therefore, a reference synergy module (RSM) is designed by averaging the muscle synergy modules for healthy subjects, and the similarities can be calculated by comparing the synergy modules for healthy subjects or patients with the RSM. In the experiment carried out in this study, average similarities of the three synergy modules for healthy subjects are respectively 0.97166, 0.87368 and 0.84932, and on the other hand, the average similarities for the three synergy modules for patients are respectively 0.59979, 0.56426 and 0.69042. Therefore, the similarities for healthy subjects are much higher than those for SCI patients, which denotes that the similarity between an individual synergy module and the RSM can be used as an objective assessing index for evaluating patients' motor function.","PeriodicalId":151112,"journal":{"name":"2022 IEEE International Conference on Robotics and Biomimetics (ROBIO)","volume":"62 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129493327","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}
In this paper we propose a multi-modal behavior planning framework for guide robots, to better assist the visually impaired to select safe paths in a cluttered space. Most prior robotic guiding systems only use physical contact, limiting their ability from operating in narrow and cluttered environments. Our multi-modal behavior planning framework is based on the Social Force Model(SFM) and the Monte Carlo Tree Search(MCTS). The proposed framework extracts robot behaviors' impact as the social force on human and predicts human motion, then employs the MCTS to search best multi-modal behavior policy. The proposed approach is deployed on a humanoid robot to guide a blind-folded person to safely travel in a complicated space.
{"title":"A Multi-modal Behavior Planning Framework for Guide Robot","authors":"Zonghao Mu, Wei Fang, Shiqiang Zhu, Tianlei Jin, Wei Song, Xiangming Xi, Qiulan Huang, J. Gu, Songyu Yuan","doi":"10.1109/ROBIO55434.2022.10011739","DOIUrl":"https://doi.org/10.1109/ROBIO55434.2022.10011739","url":null,"abstract":"In this paper we propose a multi-modal behavior planning framework for guide robots, to better assist the visually impaired to select safe paths in a cluttered space. Most prior robotic guiding systems only use physical contact, limiting their ability from operating in narrow and cluttered environments. Our multi-modal behavior planning framework is based on the Social Force Model(SFM) and the Monte Carlo Tree Search(MCTS). The proposed framework extracts robot behaviors' impact as the social force on human and predicts human motion, then employs the MCTS to search best multi-modal behavior policy. The proposed approach is deployed on a humanoid robot to guide a blind-folded person to safely travel in a complicated space.","PeriodicalId":151112,"journal":{"name":"2022 IEEE International Conference on Robotics and Biomimetics (ROBIO)","volume":"31 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124537142","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}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：