Search Results (220)

Piano fingering estimation remains a complex problem due to the combinatorial nature of hand movements and no best solution for any situation. A recent model-free reinforcement learning framework for piano fingering modeled each monophonic piece as an environment and demonstrated that value-based methods outperform probability-based approaches. Building on their finding, this paper addresses the more complex polyphonic fingering problem by formulating it as an online model-free reinforcement learning task with a novel training strategy. Thus, we introduce a novel Elite Episode Replay (EER) method to improve learning efficiency by prioritizing high-quality episodes during training. This strategy accelerates early reward acquisition and improves convergence without sacrificing fingering quality. The proposed architecture produces multiple-action outputs for polyphonic settings and is trained using both elite-guided and uniform sampling. Experimental results show that the EER strategy reduces training time per step by 21% and speeds up convergence by 18% while preserving the difficulty level and result of the generated fingerings. An empirical study of elite memory size further highlights its impact on training performance in solving piano fingering estimation. Full article

The research presented in this paper focuses on the utilization of 3D printing technology in the design and manufacture of a prosthetic hand, equipped with a digit replicator. The subject of this study was a young man who had undergone the amputation of two fingers on his right hand. The electronic control of the movement of the finger copy was developed using Arduino language. A concept and outline drawings were developed in ProCreate. Three-dimensional scan of the hand and forearm was made using an EinScan PRO HD SHINING 3D scanner. Using CAD software—Autodesk Inventor and Autodesk Meshmixer, the prosthesis was designed. Printing was carried out on a 3D printer of the i3 MK3 and MK3+ series using a PLA (polylactic acid) filament. It was determined that PLA is an optimal material for printing, as it is considered to be safe for future patients’ skin. Work on the electronic circuitry started in Autodesk TinkerCad simulation software, allowing the code to be verified and ensuring the safety of the control system. The prosthesis’s design demonstrates the potential to reach as many people in need as possible by using readily available, low-cost, and easy-to-use components. Full article

The advent of mixed reality (MR) systems has revolutionized human–computer interactions by seamlessly integrating virtual elements with the real world. Devices like the HoloLens 2 (HL2) enable intuitive, hands-free interactions through advanced hand-tracking technology, making them valuable in fields such as education, healthcare, engineering, and training simulations. However, despite the growing adoption of MR, there is a noticeable lack of comprehensive comparisons between the hand-tracking accuracy of the HL2 and high-precision benchmarks like motion capture systems. Such evaluations are essential to assess the reliability of MR interactions, identify potential tracking limitations, and improve the overall precision of hand-based input in immersive applications. This study aims to assess the accuracy of HL2 in tracking hand position and measuring kinematic hand parameters, including joint angles and lateral pinch span (distance between thumb and index fingertips), using its tracking data. To achieve this, the Vicon motion capture system (VM) was used as a gold-standard reference. Three tasks were designed: (1) finger tracing of a 2D pattern in 3D space, (2) grasping various common objects, and (3) lateral pinching of objects with varying sizes. Task 1 tests fingertip tracking, Task 2 evaluates joint angle accuracy, and Task 3 examines the accuracy of pinch span measurement. In all tasks, HL2 and VM simultaneously recorded hand positions and movements. The data captured in Task 1 were analyzed to evaluate HL2’s hand-tracking capabilities against VM. Finger rotation angles from Task 2 and lateral pinch span from Task 3 were then used to assess HL2’s accuracy compared to VM. The results indicate that the HL2 exhibits millimeter-level errors compared to Vicon’s tracking system in Task 1, spanning in a range from 2 mm to 4 mm, suggesting that HL2’s hand-tracking system demonstrates good accuracy. Additionally, the reconstructed grasping positions in Task 2 from both systems show a strong correlation and an average error of 5°, while in Task 3, the accuracy of the HL2 is comparable to that of VM, improving performance as the object thickness increases. Full article

The widespread use of WiFi Internet-of-Things (IoT) devices has rendered them valuable tools for detecting information about the physical environment. Recent studies have demonstrated that WiFi Channel State Information (CSI) can detect physical events like movement, occupancy increases, and gestures. This paper empirically investigates the conditions under which WiFi sensing technology remains effective for keystroke detection. To achieve this timely goal of assessing whether it can raise any privacy concerns, experiments are conducted using commodity hardware to predict the accuracy of WiFi CSI in detecting keys pressed on a keyboard. Our novel results show that, in an ideal setting with a robotic arm, the position of a specific key can be predicted with $99\%$ accuracy using a simple machine learning classifier. Furthermore, human finger localisation over a key and actual key-press recognition is also successfully achieved, with $94\%$ and $89\%$ reduced accuracy values, respectively. Moreover, our detailed investigation reveals that to ensure high accuracy, the gap distance between each test object must be substantial, while the size of the test group should be limited. Finally, we show WiFi sensing technology has limitations in small-scale gesture recognition for generic settings where proper device positioning is crucial. Specifically, detecting keyed words achieves an overall accuracy of $94\%$ for the forefinger and $87\%$ for multiple fingers when only the right hand is used. Accuracy drops to $56\%$ when using both hands. We conclude WiFi sensing is effective in controlled indoor environments, but it has limitations due to the device location and the limited granularity of sensing objects. Full article

Soft anthropomorphic robotic grippers are attractive because of their inherent compliance, allowing them to adapt to the shape of grasped objects and the overload protection needed for safe human–robot interaction or gripping delicate objects with sophisticated control. The anthropomorphic design allows the gripper to benefit from the biological evolution of the human hand to create a multi-functional robotic end effector. Entirely soft grippers could be more efficient because they yield under high loads. A trending solution is a hybrid gripper combining soft and rigid elements. This work describes a prototype of an anthropomorphic, underactuated five-finger gripper with a direct pneumatic drive from soft bending actuators and an integrated resistive tactile sensor array. It is a hybrid construction with soft robotic structures and rigid skeletal elements, which reinforce the body, focus the direction of the actuator’s movement, and make the finger joints follow the forward kinematics. The hand is equipped with a resistive tactile dielectric elastomer sensor array that directly triggers the hand’s actuation in the sense of reflexes. The hand can execute precision grips with two and three fingers, as well as lateral grip and strong grip types. The softness of the actuation allows the finger to adapt to the shape of the objects. Full article

This study aims to present the development of a wearable prototype device consisting of soft robotic gloves (SRGs), its integration into a wearable robotics platform for sensorimotor rehabilitation, and the device’s validation experiments with individuals suffering from impaired hand motor function due to neurological lesions. The SRG is tested and evaluated by users with spinal cord injury (SCI) and stroke. The proposed system combines multiple-sensor arrays with pneumatic actuation to assist finger movement during grasping tasks. Evaluations on SCI and stroke patients revealed that the gloves consistently improved finger and grip performance. Detailed analyses indicated observable differences in sensor-derived features during actuation versus non-actuation, with statistically significant modifications appearing in both time-domain and frequency-domain metrics. Although the stroke participants exhibited greater variability, all participants were able to use the system reporting low discomfort and effort. The findings underscore the potential for personalized calibration to further optimize therapeutic outcomes. In summary, the study validates the utility of these gloves as assistive and rehabilitative modalities, and future research will focus on refining the device in the context of multimodal wearable robotics and individualized neurorehabilitation strategies. Full article

An anthropomorphic robot is a mechanical device designed to perform human-like tasks, such as manipulating objects, and has been one of the significant contributions in robotics over the past 60 years. This paper presents an advanced system for controlling a robotic arm using user hand gestures and movements. It eliminates the need for traditional sensors or physical controls by implementing an intuitive approach based on MediaPipe and computer vision. The system recognizes the user’s hand movements. It translates them into commands that are sent to a microcontroller, which operates a robotic hand equipped with six servomotors: five for the fingers and one for the wrist, which stands out for its orthonormal design that avoids occlusion problems in turns of up to 180°, guaranteeing precise wrist control. Unlike conventional systems, this approach uses only a 2D camera to capture movements, simplifying design and reducing costs. The proposed system allows replicating the user’s activity with high precision, expanding the possibilities of human-robot interaction. Notably, the system has been able to replicate the user’s hand gestures with an accuracy of up to 95%. Full article

Background/Objectives: When manipulating an object placed on the palm, the eyes and fingers move together. To perform this task precisely, coordination of the eyes and fingers is needed. Based on this view, the present study examined the three-way interaction among the central motor drives to the eyes, index finger, and little finger. Methods: Healthy male participants abducted the right index and/or little finger with or without concomitant saccadic eye movement to the right in response to a visual cue, while the forearm was in the pronated or supinated position. We measured the reaction time (RT), velocity, and amplitude of the eye movements, as well as the RT and amplitude of the electromyographic (EMG) responses in the prime movers for the independent and dependent finger movements. Results: The velocity, amplitude, and RT of the eye movement were not changed by the additional involvement of the finger movement, indicating that the central motor drive to the finger does not influence the eye motor excitability and central motor drive to the eyes. On the one hand, the RT of the finger was not changed by the eye movement, indicating that the central motor drive to the eyes does not influence the central motor drive to the finger muscle. On the other hand, the EMG amplitude in the first dorsal interosseous muscle at the movement onset decreased during the concomitant eye movement, indicating that the central motor drive to the eyes suppresses the motor excitability of the independent finger muscle. The RT increased and EMG amplitude decreased in one finger muscle when the other finger concurrently moved, indicating that the central motor drive to one finger muscle suppresses the motor excitability of and central motor drive to the other finger muscle. The change in the RT and EMG amplitude in one finger muscle caused by the concomitant execution of the other finger movement and/or eye movement varied with forearm position, indicating that forearm proprioception influences the interaction of the motor execution processes among the fingers and eyes. Conclusions: The central motor drive to the eyes or finger muscles suppresses the motor excitability of the other finger muscles and the central motor drive to that muscle, but the central motor drive to the finger muscles does not influence those for the eyes. Forearm proprioception influences the motor excitability of the finger muscle and central motor drive to that muscle. Full article

To improve the adaptability of the hand prosthesis, we propose a new smart control for a physiological finger prosthesis using the advantages of the deep deterministic policy gradient (DDPG) algorithm. A rigid body model was developed to represent the finger as a training environment. The geometric characteristics and physiological physical properties of the finger available in the literature were assumed, but the joint’s stiffness and damping were neglected. The standard DDPG algorithm was modified to train an artificial neural network (ANN) to perform two predetermined trajectories: linear and sinusoidal. The ANN was evaluated through the use of a computational model that simulated the functionality of the finger prosthesis. The model demonstrated the capacity to successfully execute both sinusoidal and linear trajectories, exhibiting a mean error of $3.984\pm 2.899$ mm for the sinusoidal trajectory and $3.220\pm 1.419$ mm for the linear trajectory. Observing the torques, it was found that the ANN used different strategies to control the movement in order to adapt to the different trajectories. Allowing the ANN to use a combination of both trajectories, our model was able to perform trajectories that differed from purely linear and sinusoidal, showing its ability to adapt to the movement of the physiological finger. The results showed that it was possible to develop a controller for multiple trajectories, which is essential to provide more integrated and personalized prostheses. Full article

To address the limitations of single-chamber soft grippers, such as constant curvature, insufficient motion flexibility, and restricted fingertip movement, this study proposes a soft gripper inspired by the structure of the human hand. The designed soft gripper consists of three fingers, each comprising three soft joints and four phalanges. The air chambers in each joint are independently actuated, enabling flexible grasping by adjusting the joint air pressure. The constraint layer is composed of a composite material with a mass ratio of 5:1:0.75 of PDMS base, PDMS curing agent, and PTFE, which enhances the overall finger stiffness and fingertip load capacity. A nonlinear mathematical model is established to describe the relationship between the joint bending angle and actuation pressure based on the constant curvature assumption. Additionally, the kinematic model of the finger is developed using the D–H parameter method. Finite element simulations using ABAQUS analyze the effects of different joint pressures and phalange lengths on the grasping range, as well as the fingertip force under varying actuation pressures. Bending performance and fingertip force tests were conducted on the soft finger actuator, with the maximum fingertip force reaching 2.21 N. The experimental results show good agreement with theoretical and simulation results. Grasping experiments with variously sized fruits and everyday objects demonstrate that, compared to traditional single-chamber soft grippers, the proposed humanoid joint-inspired soft gripper significantly expands the grasping range and improves grasping force by four times, achieving a maximum grasp weight of 0.92 kg. These findings validate its superior grasping performance and potential for practical applications. Full article

Background/Objectives: This study evaluated the effects of a tactile-discrimination compensatory real-time feedback device on hand sensorimotor function in cervical spinal cord injury patients. The study assessed changes in hand numbness, dexterity, and electroencephalogram (EEG) activity, particularly γ-wave power in the sensorimotor area during skilled finger movements. Methods: Three patients with cervical spinal cord injury who presented with hand sensorimotor dysfunction underwent treatment with this device. All cases underwent the intervention using an AB design; A is the exercise task without the system device, and B is the exercise task under the system device. To confirm the reproducibility and minimize the influence of confounding factors, a multiple-baseline design, in which the intervention period was staggered for each subject, was applied. To determine efficacy, the hand numbness numerical rating scale, peg test, and EEG were measured daily, and Tau-U calculations were performed. Results: In two of three cases, moderate or very large changes were observed in numbness in B. In all cases, there was a large or very large change in the peg test results in the B. Regarding EEG activity, the non-skilled participants showed amplification of γ-wave power in the sensorimotor area during the B. Conversely, in the skilled participants, the γ-wave power of the sensorimotor area was attenuated during skillful movements. Conclusions: These findings indicate that the ability of the brain to compare and align predictive control with sensory feedback might be compromised in patients with damage to the afferent pathways of the central nervous system. Moreover, the use of this device appears to have played a role in supporting functional recovery. Full article

This study introduces a novel system for evaluating hand motor function through synergy-based analysis during object manipulation in virtual and mixed-reality environments. Conventional assessments of hand function are often subjective, relying on visual observation by therapists or patient-reported outcomes. To address these limitations, we developed a system that utilizes the leap motion controller (LMC) to capture finger motion data without the constraints of glove-type devices. Spatial synergies were extracted using principal component analysis (PCA) and Varimax rotation, providing insights into finger motor coordination with the sparse decomposition. Additionally, we incorporated the HoloLens 2 to create a mixed-reality object manipulation task that enhances spatial awareness for the user, improving natural interaction with virtual objects. Our results demonstrate that synergy-based analysis allows for the systematic detection of hand movement abnormalities that are not captured through traditional task performance metrics. This system demonstrates promise in advancing rehabilitation by enabling more objective and detailed evaluations of finger motor function, facilitating personalized therapy, and potentially contributing to the early detection of motor impairments in the future. Full article

Ensuring comfort in light mobility is a crucial aspect for supporting individuals’ well-being and safety while driving scooters, riding bicycles, etc. In fact, factors such as the hand grip on the handlebar, positions of the wrist and arm, overall body posture, and affecting vibrations play key roles. Wearable systems offer the ability to noninvasively monitor physiological parameters, such as body temperature and heart rate, aiding in personalized comfort assessment. In this context, user positions while driving or riding are, on the other hand, more challenging to monitor ecologically. Developing effective smart gloves as a support for comfort and movement monitoring introduces technical complexities, particularly in sensor selection and integration. Light and flexible sensors can help in this regard by ensuring reliable sensing and thus addressing the optimization of the comfort for the driver. In this work, a novel wireless smart glove is proposed, integrating four bend sensors, four force-sensitive sensors, and one inertial measurement unit for measuring the finger movements, hand orientation, and the contact force exerted by the hand while grasping the handlebar during driving or riding. The smart glove has been proven to be repeatable (1.7%) and effective, distinguishing between different grasped objects, such as a flask, a handlebar, a tennis ball, and a small box. Additionally, it proved to be a valuable tool for monitoring specific actions while riding bicycles, such as braking, and for optimizing the posture during the ride. Full article

Background/Objectives: The steady increase in the global older adult population highlights critical challenges, including the development of preventive strategies to extend healthy life expectancy and support independence in activities of daily living. Although there is an aging-related reduction in manual dexterity, the difference in bimanual coordination performance between young and older adults remains unclear. We aimed to elucidate the characteristics of bimanual coordination among young, young-old, and old-old adult participants. Methods: The participants performed in-phase (tapping the thumb and index finger together as fast as possible) and anti-phase (alternating movement between the left and right fingers) bimanual coordination tasks, and intergroup comparison of the task parameters was performed. The receiver operating characteristic curve was also conducted to calculate age cut-off points for bimanual coordination. Results: The number and frequency of taps significantly decreased sequentially in young, young-old, and old-old adults, whereas the average of tap interval significantly increased in this order (p < 0.05). There was no significant difference between the young-old and old-old groups in the average local maximum distance (p > 0.05). These findings indicate that bimanual coordination task performance varies depending on specific parameters. Furthermore, the age cut-off points for bimanual coordination were determined as 68.5 years for the right-hand number of taps (AUC = 0.73) in the anti-phase task, 73.5 years for the right-hand average of tapping interval (AUC = 0.72) in the anti-phase task, and 65.5 years for the left-hand frequency of taps (AUC = 0.72) of the anti-phase task. Conclusions: the number of taps, average of tapping interval, and frequency of taps are potential indicators of aging-related changes in bimanual coordination. Full article

The coupled-adaptive underactuated finger offers two motion modes: pre-grasping and self-adaptive grasping. It can execute anthropomorphic pre-grasp motions before the proximal phalanx contacts an object and transitions to adaptive enveloping once contact occurs. The key to designing a coupled-adaptive finger lies in its configuration and parameter, which are crucial for achieving a more human-like design for the prosthetic hand. Thus, this paper proposes a configuration topology and parameter optimization design method for a three-joint coupled-adaptive underactuated finger. The finger mechanism utilizes a combination of prismatic pairs and a compression spring to facilitate the transition between coupled motion and adaptive motion. This enables the underactuated finger to perform coupled movements in free space and adaptive grasping motions once it makes contact with an object. Furthermore, this paper introduces a finger linkage parameter optimization method that takes the joint motion angles and overall dimensions as constraints, aiming to linearize the joint coupling motion ratios as the primary optimization objective. The design method proposed in this paper not only presents a novel linkage mechanism but also outlines and compares its isomorphic types. Furthermore, the optimization results provide an accurate maximum motion value for the finger. Full article