Article
Robotics
Fatemeh Zahedi, Dongjune Chang, Hyunglae Lee
Summary: This letter introduces a user-adaptive variable damping controller that improves the overall performance of coupled human-robot systems during physical interaction. Bayesian optimization is used to evaluate and optimize the controller performance, considering the uncertainty of human behaviors and noisy observations. Experiments with a robotic arm manipulator show that the adaptive control strategy significantly reduces energy expenditure and improves stability, agility, and user effort.
IEEE ROBOTICS AND AUTOMATION LETTERS
(2022)
Article
Robotics
Benjamin A. Newman, Reuben M. Aronson, Siddhartha S. Srinivasa, Kris Kitani, Henny Admoni
Summary: The HARMONIC dataset presents a large multimodal dataset capturing human interactions with a robotic arm in a shared autonomy setting designed for assistive eating tasks. It includes data views of various aspects such as eye gaze, body pose, and hand pose, providing valuable resources for researchers interested in intention prediction, human mental state modeling, and shared autonomy studies. The dataset offers data streams in formats like video and human-readable CSV and YAML files.
INTERNATIONAL JOURNAL OF ROBOTICS RESEARCH
(2022)
Article
Chemistry, Multidisciplinary
Harsh Maithani, Juan Antonio Corrales Ramon, Laurent Lequievre, Youcef Mezouar, Matthieu Alric
Summary: This study demonstrates a proof of concept of a physical human-robot interaction-based assistive strategy for an industrial meat cutting system, which can potentially be transferred to an exoskeleton. The research focuses on how a robot can assist a human in meat processing, specifically in the meat cutting industry, by developing two assistive strategies and integrating them into an impedance controller.
APPLIED SCIENCES-BASEL
(2021)
Article
Chemistry, Analytical
Yu-Chen Sun, Meysam Effati, Hani E. E. Naguib, Goldie Nejat
Summary: When we talk about soft in the context of socially assistive robots (SARs), we usually refer to the soft outer shells of these robots. However, soft robotics, which incorporates smart materials for biomimetic motions and responsive sensing, has not been fully utilized in SAR design. This perspective article introduces SoftSAR as a new collaborative research area that explores the potential benefits of soft robotics in SARs. It discusses how soft robotic mechanisms can enhance actuation and sensory designs, and how they can inform future SAR applications.
Article
Robotics
Carly Thalman, Marielle Prescott Debeurre, Hyunglae Lee
Summary: The study found that the novel soft robotic ankle-foot orthosis (SR-AFO) has significant potential in gait entrainment, effectively improving gait adaptation and rehabilitation. All 10 subjects were successfully entrained with the SR-AFO, and the observed basin of entrainment was substantially higher compared to a heavy rigid ankle robot, confirming its effectiveness.
IEEE ROBOTICS AND AUTOMATION LETTERS
(2021)
Review
Computer Science, Information Systems
S. M. Bhagya P. Samarakoon, M. A. Viraj J. Muthugala, A. G. Buddhika P. Jayasekara
Summary: An emerging trend of using service robots in various applications is seen as a promising effort to improve the quality of life. These robots are designed for non-expert users and often need to navigate in environments with humans. Therefore, it is important for these robots to exhibit human-friendly navigation behavior. This paper presents a review on Human-Robot Proxemics (HRP), including user studies and methods for establishing HRP awareness in service robots. The review identifies limitations in current state-of-the-art research and suggests potential future work. Additionally, it summarizes important HRP parameters and behavior from existing user studies, providing valuable data for developing HRP-aware behavior in service robots.
Article
Chemistry, Analytical
Giorgia Chiriatti, Luca Carbonari, Maria Gabriella Ceravolo, Elisa Andrenelli, Marzia Millevolte, Giacomo Palmieri
Summary: One of the most interesting characteristics of collaborative robots is their ability to be used in close cooperation scenarios. In industry, this facilitates the implementation of human-in-loop workflows. This paper presents a rehabilitation framework for the upper limbs of neurological patients, utilizing a collaborative robot to guide and improve patients' coordination through three-dimensional trajectories. Experimental results from 19 volunteers showed positive feedback, with 52% of subjects willing to return and 44% enjoying the exercise. Patients were able to perform the exercise with a maximum deviation of 16 mm and limited muscular effort (average maximum forces recorded at around 50 N).
Review
Robotics
Benjamin A. Newman, Reuben M. Aronson, Kris Kitani, Henny Admoni
Summary: The article introduces a definition of assistance, surveys existing work, and explores three key design axes that occur in many domains. It emphasizes the importance of understanding a robot in control of its actions subordinate to a user's goals.
FRONTIERS IN ROBOTICS AND AI
(2022)
Article
Robotics
Aleksandra Kalinowska, Millicent Schlafly, Kyra Rudy, Julius P. A. Dewald, Todd D. Murphey
Summary: This study proposes a frequency-based assessment method to measure the bandwidth of physical human-robot interaction. The results show that haptic cues improve motion timing and performance in dynamic tasks compared to visual-only cuing. This method is important for assistive robots to adapt to individuals and for assessment and physical rehabilitation purposes.
IEEE ROBOTICS AND AUTOMATION LETTERS
(2022)
Article
Chemistry, Analytical
Enrique Coronado, Toshifumi Shinya, Gentiane Venture
Summary: In recent years, there has been a growing interest in developing robotic systems to improve the quality of life for individuals of all ages. This article introduces a novel system architecture that enables the Pepper robot to walk side-by-side while holding hands and communicate with the surrounding environment. The system achieves this by estimating the applied force using an observer and utilizing object recognition for communication. The integration of these components demonstrates the system's capability to achieve its intended purpose.
Article
Computer Science, Information Systems
Roman Michalik, Ales Janota, Michal Gregor, Marian Hrubos
Summary: This paper discusses an approach for controlling a cooperating YuMi robot using hand gestures recognized by a camera and artificial intelligence through a TCP/IP connection with Python. The program can be enhanced by integrating other IoT devices for robot control and data collection for specific applications.
Article
Robotics
Yves Zimmermann, Jaeyong Song, Cedric Deguelle, Julia Laderach, Lingfei Zhou, Marco Hutter, Robert Riener, Peter Wolf
Summary: Exoskeleton robots have found applications in various areas such as neurorehabilitation, telemanipulation, and power augmentation. However, existing attachment concepts have imposed limitations on the performance of modern exoskeletons due to issues like insufficient stiffness, time-consuming adaption processes, and bulkiness. In this study, an augmented attachment system was proposed and compared to a conventional solution. The results showed that the proposed system significantly reduced the relative motion between the human and the robot, mitigated undesired contacts, and facilitated better alignment with the user's anatomy.
IEEE TRANSACTIONS ON ROBOTICS
(2023)
Article
Robotics
Xiaowei Tan, Bi Zhang, Guangjun Liu, Xingang Zhao, Yiwen Zhao
Summary: This article introduces a time-independent exoskeleton control system that enables users to retain full control over the assistance provided by the exoskeleton. The control system utilizes a novel assistance profile generation method and an iterative force control method to continuously adjust the assistance based on the user's movement. Experimental results demonstrate the effectiveness of the proposed control system compared to a common time-dependent control system.
IEEE TRANSACTIONS ON ROBOTICS
(2023)
Article
Robotics
Zackory Erickson, Henry M. Clever, Vamsee Gangaram, Eliot Xing, Greg Turk, C. Karen Liu, Charles C. Kemp
Summary: This study introduces a capacitive servoing control scheme that allows a robot to accurately sense and navigate around human limbs during close physical interactions. The results indicate that multidimensional capacitive servoing enables a robot's end effector to move proximally or distally along human limbs while adapting to human pose.
IEEE TRANSACTIONS ON ROBOTICS
(2023)
Review
Agriculture, Multidisciplinary
George Adamides, Yael Edan
Summary: Despite the active research and development in agricultural robotics in the past decade, practical implementation of these robots is still limited due to the complex agricultural environment. Human-robot collaboration is suggested as a potential solution to enhance performance. This study provides a quantitative review of 27 distinct projects conducted over the past 21 years, revealing that important topics in agricultural human-robot collaboration focus on safety, ergonomics, awareness, and productivity, with a focus on harvesting and spraying tasks. Furthermore, a roadmap is proposed to define the research and development directions necessary to improve performance and implement human-robot collaboration in agriculture, including aspects of efficient collaboration, functionality, performance measures, and experimentation procedures. Constraints, future challenges, and prospects are also discussed.
COMPUTERS AND ELECTRONICS IN AGRICULTURE
(2023)
Article
Chemistry, Analytical
Manuel Fernandez-Carmona, Joaquin Ballesteros, Marta Diaz-Boladeras, Xavier Parra-Llanas, Cristina Urdiales, Jesus Manuel Gomez-de-Gabriel
Summary: Smart rollators can autonomously capture gait parameters and avoid complex setups, with the introduction of a low cost open-source modular rollator in this work. The system is based on commercial components and its software architecture runs over ROS2, allowing for further customization and expansion.
Article
Geochemistry & Geophysics
Man Zhou, Jie Huang, Feng Zhao, Danfeng Hong
Summary: This study proposes a novel information compensation and integration network for pan-sharpening, addressing the mutual dependencies between PAN and MS modalities through effective cross-modality joint learning. The method employs cross-central difference convolution and classical back-projection technique to extract texture details and guide the intrinsic information learning of MS images iteratively. It also uses a hierarchical transformer to integrate stage-iteration information from spatial and temporal contexts. Experimental results demonstrate the superiority of the proposed method over existing state-of-the-art methods on multiple satellite datasets.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2023)
Article
Computer Science, Artificial Intelligence
Wang Yuan, Yong-Hua Liu, Chun-Yi Su, Feng Zhao
Summary: This article proposes a whole-body control approach for a nonholonomic mobile manipulator using model predictive control (MPC) and fuzzy logic system (FLS). The strategy integrates an MPC-based WBC strategy with an adaptive FLS to approximate the unknown dynamics and provide optimal control inputs. Experimental results demonstrate the effectiveness of the proposed method in trajectory tracking.
IEEE TRANSACTIONS ON FUZZY SYSTEMS
(2023)
Article
Computer Science, Artificial Intelligence
Zongzhe Sun, Feng Zhao
Summary: Visible-infrared person re-identification (VI-ReID) tackles cross-modality matching between daytime visible and nighttime infrared images. Existing methods enhance multi-modality feature representations using attention modules, but overlook attention quality and lack effective supervision of the attention learning process. To address these issues, a counter-factual attention alignment (CAA) strategy is proposed, which mines intra-modality attention information and aligns cross-modality attentions. Experimental results show that the proposed approach outperforms other state-of-the-art methods on two standard benchmarks.
PATTERN RECOGNITION LETTERS
(2023)
Article
Computer Science, Artificial Intelligence
Jiannan Zhao, Hongxin Wang, Nicola Bellotto, Cheng Hu, Jigen Peng, Shigang Yue
Summary: Collision detection is a challenging task for unmanned aerial vehicles, especially for small or micro-UAVs with limited computational power. Insects like locusts demonstrate remarkable ability to avoid collision through their visual neuron called LGMD. However, existing LGMD models cannot distinguish looming clearly from other visual cues. To address this issue, a new model based on distributed spatial-temporal synaptic interactions is proposed, which has been shown to enhance looming selectivity in complex flying scenes. This model has the potential to be implemented on embedded collision detection systems for small or micro-UAVs.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2023)
Article
Robotics
Tomasz Piotr Kucner, Martin Magnusson, Sariah Mghames, Luigi Palmieri, Francesco Verdoja, Chittaranjan Srinivas Swaminathan, Tomas Krajnik, Erik Schaffernicht, Nicola Bellotto, Marc Hanheide, Achim J. Lilienthal
Summary: Robotic mapping provides spatial information for autonomous agents, ranging from simple 2D representations to complex semantic maps. This survey article focuses on maps of dynamics (MoDs), which store semantic information about motion patterns. MoDs can be used for motion planning, localization, and human motion prediction. The survey organizes knowledge in the field, introduces a novel taxonomy, and identifies promising directions for future work. The field is mature, with increasing use of MoDs, but still has potential for significant impact through novel contributions.
INTERNATIONAL JOURNAL OF ROBOTICS RESEARCH
(2023)
Article
Engineering, Multidisciplinary
Marta Diaz-Boladeras, Xavier Llanas, Manuel Fernandez-Carmona, Andreu Catala, Cristina Urdiales, Marta Muste, Elsa Perez, Alex Barco
Summary: In assisted ambulation, the user's psychological comfort plays a significant role in the acceptability of mobility aids and overall gait performance. This study investigated the effect of distress on rollator assisted navigation and proposed a novel test protocol using the Self-Assessment Manikin questionnaire. Preliminary results showed that stressors like uneven or sloping surfaces increased perceived stress, while the shape of the trajectory did not significantly affect stress.
COGENT ENGINEERING
(2023)
Article
Automation & Control Systems
Sergi Molina, Anna Mannucci, Martin Magnusson, Daniel Adolfsson, Henrik Andreasson, Mazin Hamad, Saeed Abdolshah, Ravi Teja Chadalavada, Luigi Palmieri, Timm Linder, Chittaranjan Srinivas Swaminathan, Tomasz Piotr Kucner, Marc Hanheide, Manuel Fernandez-Carmona, Grzegorz Cielniak, Tom Duckett, Federico Pecora, Simon Bokesand, Kai O. Arras, Sami Haddadin, Achim J. Lilienthal
Summary: Current intralogistics services aim to meet the demands of e-commerce, minimize delivery times and waste, and enhance overall flexibility. As a result, the adoption of automated guided vehicles (AGVs) and, more recently, autonomous mobile robots (AMRs) is steadily increasing in logistics operations.
IEEE ROBOTICS & AUTOMATION MAGAZINE
(2023)
Article
Computer Science, Information Systems
Jie Huang, Xueyang Fu, Zeyu Xiao, Feng Zhao, Zhiwei Xiong
Summary: In order to improve the image quality of stereo cameras under low-light conditions, a novel Dual-View Enhancement Network (DVENet) based on the Retinex theory is proposed. It consists of two stages: the first stage estimates an illumination map to enhance the correlation of two views, while the second stage integrates information from two views to recover details with the guidance of the illumination map. Experimental results demonstrate the superiority of the proposed method over existing solutions on both synthetic and real-world stereo datasets.
IEEE TRANSACTIONS ON MULTIMEDIA
(2023)
Article
Geochemistry & Geophysics
Kaiwen Zheng, Jie Huang, Man Zhou, Danfeng Hong, Feng Zhao
Summary: In this work, we propose an uncertainty-aware adaptive pansharpening network (UAPN) that integrates PAN information spatial variantly to restore LMS information with an uncertainty mechanism. Experimental results demonstrate the superiority of our UAPN with fewer parameters and FLOPs, outperforming other state-of-the-art methods both qualitatively and quantitatively on multiple satellite datasets.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2023)
Article
Computer Science, Artificial Intelligence
Man Zhou, Jie Huang, Danfeng Hong, Feng Zhao, Chongyi Li, Jocelyn Chanussot
Summary: This paper proposes a closed-loop scheme that learns both the pan-sharpening and its corresponding degradation process simultaneously to regulate the solution space. A specific multiscale high-frequency texture extraction module is also designed to strengthen the algorithm. Experimental results demonstrate the effectiveness of the proposed algorithm.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2023)
Proceedings Paper
Computer Science, Artificial Intelligence
Manuel Fernandez-Carmona, Gabriela Verdezoto, Joaquin Ballesteros, Jesus M. Gomez-de-Gabriel, Cristina Urdiales
Summary: The increasing ratio of elderly people per caregiver calls for changes in healthcare systems to provide enhanced, value-added and cost-effective services. Industry is already utilizing technologies such as AI, deep learning, data analytics and robotics in this process. Health 4.0, a parallel approach, opens up new opportunities. Smart assistive devices, like smart rollators, can gather data to provide user condition services for caregivers to assess gait abnormalities and functional fitness conditions.
PROCEEDINGS OF THE INTERNATIONAL CONFERENCE ON UBIQUITOUS COMPUTING & AMBIENT INTELLIGENCE (UCAMI 2022)
(2023)
Article
Computer Science, Information Systems
Naishan Zheng, Jie Huang, Feng Zhao, Xueyang Fu, Feng Wu
Summary: This paper proposes an Unsupervised Underexposed Image Enhancement Network (U2IENet) with self-illuminated and perceptual guidance. It can adaptively adjust the illumination to match the image-specific lightness and refine the details perceptually.
IEEE TRANSACTIONS ON MULTIMEDIA
(2023)
Proceedings Paper
Computer Science, Artificial Intelligence
Man Zhou, Jie Huang, Keyu Yan, Hu Yu, Xueyang Fu, Aiping Liu, Xian Wei, Feng Zhao
Summary: This paper proposes a Spatial-Frequency Information Integration Network (SFIIN) for pan-sharpening. The network addresses pan-sharpening in both spatial and frequency domains, and consists of three key components for integrating and learning information. Experiments demonstrate the superior performance of the proposed network compared to other methods.
COMPUTER VISION - ECCV 2022, PT XVIII
(2022)
Article
Geochemistry & Geophysics
Man Zhou, Jie Huang, Xueyang Fu, Feng Zhao, Danfeng Hong
Summary: This study proposes a heterogeneous knowledge-distilling pan-sharpening framework to enhance the processing of ground-truth multispectral images by mimicking the ground-truth reconstruction task in both the feature space and network output. Experimental results demonstrate that the proposed network outperforms state-of-the-art methods both visually and quantitatively.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2022)