Explore the vast range of titles available.

As one of the most important senses in human beings, touch can also help robots better perceive and adapt to complex environmental information, improving their autonomous decision-making and execution capabilities. Compared to other perception methods, tactile perception needs to handle multi-channel tactile signals simultaneously, such as pressure, bending, temperature, and humidity. However, directly transferring deep learning algorithms that work well on temporal signals to tactile signal tasks does not effectively utilize the physical spatial connectivity information of tactile sensors. In this paper, we propose a tactile perception framework based on graph attention networks, which incorporates explicit and latent relation graphs. This framework can effectively utilize the structural information between different tactile signal channels. We constructed a tactile glove and collected a dataset of pressure and bending tactile signals during grasping and holding objects, and our method achieved 89.58% accuracy in object tactile signal classification. Compared to existing time-series signal classification algorithms, our graph-based tactile perception algorithm can better utilize and learn sensor spatial information, making it more suitable for processing multi-channel tactile data. Our method can serve as a general strategy to improve a robot’s tactile perception capabilities.

The absence of tactile perception limits the dexterity of a prosthetic hand and its acceptance by amputees. Recreating the sensing properties of the skin using a flexible tactile sensor could have profound implications for prosthetics, whereas existing tactile sensors often have limited functionality with cross‐interference. In this study, we propose a machine‐learning‐assisted multifunctional tactile sensor for smart prosthetics, providing a human‐like tactile sensing approach for amputations. This flexible sensor is based on a poly(3,4‐ethylenedioxythiophene): poly(styrene sulfonate) (PEDOT:PSS)–melamine sponge, which enables the detection of force and temperature with low cross‐coupling owing to two separate sensing mechanisms: the open‐circuit voltage of the sensor as a force‐insensitive intrinsic variable to measure the absolute temperature and the resistance as a temperature‐insensitive extrinsic variable to measure force. Furthermore, by analyzing the unsteady heat conduction and characterizing it using real‐time thermal imaging, we demonstrated that the process of open‐circuit voltage variation resulting from the unsteady heat conduction is closely correlated with the heat‐conducting capabilities of materials, which can be utilized to discriminate between substances. Assisted by the decision tree algorithm, the device is endowed with thermal conductivity sensing ability, which allows it to identify 10 types of substances with an accuracy of 94.7%. Furthermore, an individual wearing an advanced myoelectric prosthesis equipped with the above sensor can sense pressure, temperature, and recognize different materials. We demonstrated that our multifunctional tactile sensor provides a new strategy to help amputees feel force, temperature and identify the material of objects without the aid of vision. image

In order to improve the accuracy of manipulator operation, it is necessary to install a tactile sensor on the manipulator to obtain tactile information and accurately classify a target. However, with the increase in the uncertainty and complexity of tactile sensing data characteristics, and the continuous development of tactile sensors, typical machine-learning algorithms often cannot solve the problem of target classification of pure tactile data. Here, we propose a new model by combining a convolutional neural network and a residual network, named ResNet10-v1. We optimized the convolutional kernel, hyperparameters, and loss function of the model, and further improved the accuracy of target classification through the K-means clustering method. We verified the feasibility and effectiveness of the proposed method through a large number of experiments. We expect to further improve the generalization ability of this method and provide an important reference for the research in the field of tactile perception classification.

In order to endow robots with human-like abilities to characterize and identify objects, they must be provided with tactile sensors and intelligent algorithms to select, control, and interpret data from useful exploratory movements. Humans make informed decisions on the sequence of exploratory movements that would yield the most information for the task, depending on what the object may be and prior knowledge of what to expect from possible exploratory movements. This study is focused on texture discrimination, a subset of a much larger group of exploratory movements and percepts that humans use to discriminate, characterize, and identify objects. Using a testbed equipped with a biologically inspired tactile sensor (the BioTac), we produced sliding movements similar to those that humans make when exploring textures. Measurement of tactile vibrations and reaction forces when exploring textures were used to extract measures of textural properties inspired from psychophysical literature (traction, roughness, and fineness). Different combinations of normal force and velocity were identified to be useful for each of these three properties. A total of 117 textures were explored with these three movements to create a database of prior experience to use for identifying these same textures in future encounters. When exploring a texture, the discrimination algorithm adaptively selects the optimal movement to make and property to measure based on previous experience to differentiate the texture from a set of plausible candidates, a process we call Bayesian exploration. Performance of 99.6% in correctly discriminating pairs of similar textures was found to exceed human capabilities. Absolute classification from the entire set of 117 textures generally required a small number of well-chosen exploratory movements (median = 5) and yielded a 95.4% success rate. The method of Bayesian exploration developed and tested in this paper may generalize well to other cognitive problems.

A brain computer interface (BCI) allows users to control external devices using non-invasive brain recordings, such as electroencephalography (EEG). We developed and tested a novel electrotactile BCI prototype based on somatosensory event-related potentials (sERP) as control signals, paired with a tactile attention task as a control paradigm. A novel electrotactile BCI comprises commercial EEG device, an electrical stimulator and custom software for EEG recordings, electrical stimulation control, synchronization between devices, signal processing, feature extraction, selection, and classification. We tested a novel BCI control paradigm based on tactile attention on a sensation at a target stimulation location on the forearm. Tactile stimuli were electrical pulses delivered at two proximal locations on the user's forearm for stimulating branches of radial and median nerves, with equal probability of the target and distractor stimuli occurrence, unlike in any other ERP-based BCI design. We proposed a compact electrical stimulation electrodes configuration for delivering electrotactile stimuli (target and distractor) using 2 stimulation channels and 3 stimulation electrodes. We tested the feasibility of a single EEG channel BCI control, to determine pseudo-online BCI performance, in ten healthy subjects. For optimizing the BCI performance we compared the results for two classifiers, sERP averaging approaches, and novel dedicated feature extraction/selection methods cross-validation procedures. We achieved a single EEG channel BCI classification accuracy in the range of 75.1 to 88.1% for all subjects. We have established an optimal combination of: single trial averaging to obtain sERP, feature extraction/selection methods and classification approach. The obtained results demonstrate that a novel electrotactile BCI paradigm with equal probability of attended (target) and unattended (distractor) stimuli and proximal stimulation sites is feasible. This method may be used to drive restorative BCIs for sensory retraining in stroke or brain injury, or assistive BCIs for communication in severely disabled users.

Tactile sensing is increasingly vital in robotics, especially for tasks like object manipulation and texture classification. Among tactile technologies, optical and electrical sensors are widely used, yet no rigorous direct comparison of their performance has been conducted. This paper addresses that gap by presenting a comparative study between a high-resolution optical tactile sensor (a modified TacTip) and a low-resolution electrical sensor combining accelerometers and piezoelectric elements. We evaluate both sensor types on two tasks: texture classification and coefficient of dynamic friction prediction. Various configurations and resolutions were explored, along with multiple machine learning classifiers to determine optimal performance. The optical sensor achieved 99.9% accuracy on a challenging texture dataset, significantly outperforming the electrical sensor, which reached 82%. However, for dynamic friction prediction, both sensors performed comparably, with only a 5~% accuracy difference. We also found that the optical sensor retained high classification accuracy even when image resolution was reduced to 25% of its original size, suggesting that ultra-high resolution is not essential. In conclusion, the optical sensor is the better choice when high accuracy is required. However, for low-cost or computationally efficient systems, the electrical sensor provides a practical alternative with competitive performance in some tasks.

In robotics, tactile perception is important for fine control using robot grippers and hands. To effectively incorporate tactile perception in robots, it is essential to understand how humans use mechanoreceptors and proprioceptors to perceive texture. Thus, our study aimed to investigate the impact of tactile sensor arrays, shear force, and the positional information of the robot’s end effector on its ability to recognize texture. A deep learning network was employed to classify tactile data from 24 different textures that were explored by a robot. The input values of the deep learning network were modified based on variations in the number of channels of the tactile signal, the arrangement of the tactile sensor, the presence or absence of shear force, and the positional information of the robot. By comparing the accuracy of texture recognition, our analysis revealed that tactile sensor arrays more accurately recognized the texture compared to a single tactile sensor. The utilization of shear force and positional information of the robot resulted in an improved accuracy of texture recognition when using a single tactile sensor. Furthermore, an equal number of sensors placed in a vertical arrangement led to a more accurate distinction of textures during exploration when compared to sensors placed in a horizontal arrangement. The results of this study indicate that the implementation of a tactile sensor array should be prioritized over a single sensor for enhanced accuracy in tactile sensing, and the use of integrated data should be considered for single tactile sensing.

This paper proposed a novel tactile-stimuli P300 paradigm for Brain-Computer Interface (BCI), which potentially targeted at people with less learning ability or difficulty in maintaining attention. The new paradigm using only two types of stimuli was designed, and different targets were distinguished by frequency and spatial information. The classification algorithm was developed by introducing filters for frequency bands selection and conducting optimization with common spatial pattern (CSP) on the tactile evoked EEG signals. It features a combination of spatial and frequency information, with the spatial information distinguishing the sites of stimuli and frequency information identifying target stimuli and disturbances. We investigated both electrical stimuli and vibration stimuli, in which only one target site was stimulated in each block. The results demonstrated an average accuracy of 94.88% for electrical stimuli and 95.21% for vibration stimuli, respectively.

The rich set of mechanoreceptors found in human skin 1 , 2 offers a versatile engineering interface for transmitting information and eliciting perceptions 3 , 4 , potentially serving a broad range of applications in patient care 5 and other important industries 6 , 7 . Targeted multisensory engagement of these afferent units, however, faces persistent challenges, especially for wearable, programmable systems that need to operate adaptively across the body 8 , 9 , 10 – 11 . Here we present a miniaturized electromechanical structure that, when combined with skin as an elastic, energy-storing element, supports bistable, self-sensing modes of deformation. Targeting specific classes of mechanoreceptors as the basis for distinct, programmed sensory responses, this haptic unit can deliver both dynamic and static stimuli, directed as either normal or shear forces. Systematic experimental and theoretical studies establish foundational principles and practical criteria for low-energy operation across natural anatomical variations in the mechanical properties of human skin. A wireless, skin-conformable haptic interface, integrating an array of these bistable transducers, serves as a high-density channel capable of rendering input from smartphone-based 3D scanning and inertial sensors. Demonstrations of this system include sensory substitution designed to improve the quality of life for patients with visual and proprioceptive impairments. Inspired by the art of kirigami, a haptic device based on a miniaturized electromechanical structure combined with skin as an elastic, energy-storing element demonstrates bioelastic state recovery and can be used in sensory substitution.

Deep‐sea exploration remains a challenging task as the extreme hydrostatic pressure environment, darkness, and suspended sediment launch severely hinder the capability of deep‐sea vehicles. As a complement to underwater camera, tactile perception becomes especially important in situations where machine vision is limited. However, tactile sensors utilized in deep sea, which should be able to detect pressure changes of only hundreds of pascals under high hydrostatic pressure, are still lacking. To tackle the challenge imposed by hydrostatic pressure, a simulated deep‐sea environment flexible sensor (SDEFS) is proposed, consisting of a force sensor array and a bending sensor based on hydrogels for tactile sensing in 50 MPa hydrostatic pressure environments. The force sensor is unaffected by the hydrostatic pressure and achieves high sensitivity of 82.62 N−1 under 100 MPa hydrostatic pressure. The SDEFS is utilized to classify objects based on the difference in hardness. It can accurately classify seven objects on the ground, and three objects in an underwater environment with hydrostatic pressure of 50 MPa, with total recognition accuracies of 98.3% and 96%, respectively. With high force measurement sensitivity and accurate recognition ability under water, the SDEFS is expected to provide very valuable haptic sensing and feedback in deep‐sea exploration. A deep‐sea flexible sensor, consisting of a force sensor array and a bending sensor based on hydrogels, is proposed for tactile sensing in 50 MPa hydrostatic pressure environments. The sensor is expected to provide very valuable haptic sensing and feedback in deep‐sea exploration, resource exploitation, environmental protection, etc.