Touch and vision go hand in hand, mutually enhancing our ability to understand the world. From a research perspective, the problem of mixing touch and vision together is underexplored and presents interesting challenges. To this end, we propose Tactile-Informed 3DGS, a novel approach that incorporates contact data (local depth maps) with multi-view images to achieve surface reconstruction and novel view synthesis. Our method optimises 3D Gaussian primitives to accurately model the object's geometry at points of contact. By creating a framework that decreases the transmittance at touch locations, we achieve a refined surface reconstruction, ensuring a uniformly smooth depth map. Touch is particularly useful when considering non-Lambertian objects, such as shiny or reflective surfaces, since contemporary methods tend to fail to reconstruct specular highlights with fidelity. By combining vision and tactile sensing, we achieve more accurate geometry reconstructions with fewer images than prior methods. We conduct evaluation on objects with glossy and reflective surfaces and demonstrate improved reconstruction quality in both the virtual and real world

This paper introduces an anthropomorphic robot hand built entirely using LEGO MINDSTORMS: the Educational SoftHand-A, a tendon-driven, highly-underactuated robot hand based on the Pisa/IIT SoftHand and related hands. To be suitable for an educational context, the design is constrained to use only standard LEGO pieces with tests using common equipment available at home. The hand features dual motors driving an agonist/antagonist opposing pair of tendons on each finger, which are shown to result in reactive fine control. The finger motions are synchonized through soft synergies, implemented with a differential mechanism using clutch gears. Altogether, this design results in an anthropomorphic hand that can adaptively grasp a broad range of objects using a simple actuation and control mechanism. Since the hand can be constructed from LEGO pieces and uses state-of-the-art design concepts for robotic hands, it has the potential to educate and inspire children to learn about the frontiers of modern robotics.

In recent years, researchers have integrated the historically separate, reinforcement learning (RL), and evidence-accumulation-to-bound approaches to decision modeling. A particular outcome of these efforts has been the RL-DDM, a model that combines value learning through reinforcement with a diffusion decision model (DDM). While the RL-DDM is a conceptually elegant extension of the original DDM, it faces a similar problem to the DDM in that it does not scale well to decisions with more than two options. Furthermore, in its current form, the RL-DDM lacks flexibility when it comes to adapting to rapid, context-cued changes in the reward environment. The question of how to best extend combined RL and DDM models so they can handle multiple choices remains open. Moreover, it is currently unclear how these algorithmic solutions should map to neurophysical processes in the brain, particularly in relation to so-called go/no-go-type models of decision making in the basal ganglia. Here, we propose a solution that addresses these issues by combining a previously proposed decision model based on the multichoice sequential probability ratio test (MSPRT), with a dual-pathway model of decision threshold learning in the basal ganglia region of the brain. Our model learns decision thresholds to optimize the trade-off between time cost and the cost of errors and so efficiently allocates the amount of time for decision deliberation. In addition, the model is context dependent and hence flexible to changes to the speed-accuracy trade-off (SAT) in the environment. Furthermore, the model reproduces the magnitude effect, a phenomenon seen experimentally in value-based decisions and is agnostic to the types of evidence and so can be used on perceptual decisions, value-based decisions, and other types of modeled evidence. The broader significance of the model is that it contributes to the active research area of how learning systems interact by linking the previously separate models of RL-DDM to dopaminergic models of motivation and risk taking in the basal ganglia, as well as scaling to multiple alternatives.

This paper presents a shear-based control scheme for grasping and manipulating delicate objects with a Pisa/IIT anthropomorphic SoftHand equipped with soft biomimetic tactile sensors on all five fingertips. These ‘microTac’ tactile sensors are miniature versions of the TacTip vision-based tactile sensor, and can extract precise contact geometry and force information at each fingertip for use as feedback into a controller to modulate the grasp while a held object is manipulated. Using a parallel processing pipeline, we asynchronously capture tactile images and predict contact pose and force from multiple tactile sensors. Consistent pose and force models across all sensors are developed using supervised deep learning with transfer learning techniques. We then develop a grasp control framework that uses contact force feedback from all fingertip sensors simultaneously, allowing the hand to safely handle delicate objects even under external disturbances. This control framework is applied to several grasp-manipulation experiments: first, retaining a flexible cup in a grasp without crushing it under changes in object weight; second, a pouring task where the center of mass of the cup changes dynamically; and third, a tactile-driven leader-follower task where a human guides a held object. These manipulation tasks demonstrate more human- like dexterity with underactuated robotic hands by using fast reflexive control from tactile sensing.

Endowing robots with advanced tactile abilities based on biomimicry involves designing human-like tactile sensors, computational models, and motor control policies to enhance contact information retrieval. Here, we consider compliance discrimination with a soft biomimetic tactile optical sensor (TacTip). In previous work, we proposed a vision-based approach derived from a computational model of human tactile perception to discriminate object compliance with the TacTip, based on contact area spread computation over the indenting force. In this work, we first increased the robustness of our vision-based method with a more precise estimation of the initial contact area condition, which enables correct compliance estimation also when the probing direction is other than normal to the specimen surface. Then, we integrated within our validated framework the mechanisms of internal muscular regulation (co-contraction) that humans adopt during object compliance probing, to maximize the information uptake. To this aim, we used human co-contraction patterns extracted during object softness probing to control a Variable Stiffness Actuator (that emulates the agonistic-antagonistic behavior of human muscles), which is used to actuate the indenter system endowed with the TacTip for object compliance exploration. We found that our model-based approach for compliance discrimination, fed with more precisely estimated initial conditions, significantly improves with the human-inspired impedance regulation, with respect to the usage of a rigid actuator.

THE sense of touch is an indispensable requirement for humans to effectively interact with the physical world around them and perform dexterous tasks. Similarly, this should be no different for robots. Imagine, for example, a robot that can open a bottle of medicine and dispense pills to an elderly person. Although this might seem a straightforward task for a human, it remains a significant challenge for a robot. Critically, the completion of the task depends on tactile sensing: the robot needs to receive and interpret the feedback from interacting with the bottle, determine the appropriate force based on the size and hardness of the pills, and adjust its pose to safely dispense them. Each step involves contact-rich interactions that can only be effectively deciphered through tactile sensing. Typically, tactile sensing works in conjunction with other modalities, such as vision, enabling the robot to adjust its actions dynamically and complete the task. In response to this vision of robots interacting with the physical world through touch, tactile robotics has now emerged as a key research area. Tactile robots can be defined as intelligent systems equipped with tactile sensors that can extract and process tactile data to guide their operations and interactions. The development of tactile robots presents scientific challenges, ranging from the design and fabrication of tactile sensors to methodologies for processing tactile data, integrating tactile feedback into task execution, and combining it with other sensory modalities to improve robot perception. As a result, tactile robotics demands collaborative efforts across several disciplines, involving material and data scientists …

In this paper, we present the design and benchmark of an innovative sensor, ViTacTip, which fulfills the demand for advanced multi-modal sensing in a compact design. A notable feature of ViTacTip is its transparent skin, which incorporates a ‘see-through-skin’ mechanism. This mechanism aims at capturing detailed object features upon contact, significantly improving both vision-based and proximity perception capabilities. In parallel, the biomimetic tips embedded in the sensor's skin are designed to amplify contact details, thus substantially augmenting tactile and derived force perception abilities. To demonstrate the multi-modal capabilities of ViTacTip, we developed a multi-task learning model that enables simultaneous recognition of hardness, material, and textures. To assess the functionality and validate the versatility of ViTacTip, we conducted extensive benchmarking experiments, including object recognition, contact point detection, pose regression, and grating identification. To facilitate seamless switching between various sensing modalities, we employed a Generative Adversarial Network (GAN)-based approach. This method enhances the applicability of the ViTacTip sensor across diverse environments by enabling cross-modality interpretation.

In this paper, we present the TacShade: a newly designed 3D-printed soft optical tactile sensor. The sensor is developed for shape reconstruction under the inspiration of sketch drawing that uses the density of sketch lines to draw light and shadow, resulting in the creation of a 3D-view effect. TacShade, building upon the strengths of the TacTip, a single-camera tactile sensor of large in-depth deformation and being sensitive to edge and surface following, improves the structure in that the markers are distributed within the gap of papillae pins. Variations in light, dark and grey effects can be generated inside the sensor under the external contact interactions. The contours of the contacting objects are outlined by white markers, while the contact depth characteristics can be indirectly obtained from the distribution of black pins and white markers, creating a 2.5D visualization. Based on the imaging effect, we improve the Shape from Shading (SFS) algorithm to process tactile images, enabling a coarse but fast reconstruction for the contact objects. Two experiments are performed. The first verifies TacShade’s ability to reconstruct the shape of the contact objects through one image for object distinction. The second experiment shows the shape reconstruction capability of TacShade for a large panel with ridged patterns based on the location of robots and image splicing technology.

Tactile sensing is significant for robotics since it can obtain physical contact information during manipulation. To capture multimodal contact information within a compact framework, we designed a novel sensor called ViTacTip, which seamlessly integrates both tactile and visual perception capabilities into a single, integrated sensor unit. ViTacTip features a transparent skin to capture fine features of objects during contact, which can be known as the see-through-skin mechanism. In the meantime, the biomimetic tips embedded in ViTacTip can amplify touch motions during tactile perception. For comparative analysis, we also fabricated a ViTac sensor devoid of biomimetic tips, as well as a TacTip sensor with opaque skin. Furthermore, we develop a Generative Adversarial Network (GAN)-based approach for modality switching between different perception modes, effectively alternating the emphasis between vision and tactile perception modes. We conducted a performance evaluation of the proposed sensor across three distinct tasks: i) grating identification, ii) pose regression, iii) contact localization and force estimation. In the grating identification task, ViTacTip demonstrated an accuracy of 99.72%, surpassing TacTip, which achieved 94.60%. It also exhibited superior performance in both pose and force estimation tasks with the minimum error of 0.08 mm and 0.03N, respectively, in contrast to ViTac’s 0.12 mm and 0.15N. Results indicate that ViTacTip outperforms single-modality sensors.

Imitation learning for robot dexterous manipulation, especially with a real robot setup, typically requires a large number of demonstrations. In this paper, we present a data-efficient learning from demonstration framework which exploits the use of rich tactile sensing data and achieves fine bimanual pinch grasping. Specifically, we employ a convolutional autoencoder network that can effectively extract and encode high-dimensional tactile information. Further, we develop a framework that achieves efficient multi-sensor fusion for imitation learning, allowing the robot to learn contact-aware sensorimotor skills from demonstrations. The ablation studies on encoded tactile features highlighted the effectiveness of incorporating rich contact information, which enabled dexterous bimanual grasping with active contact searching. Extensive experiments demonstrated the robustness of the fine pinch grasp policy directly learned from few-shot demonstration, including grasping of the same object with different initial poses, generalizing to ten unseen new objects, robust and firm grasping against external pushes, as well as contact-aware and reactive re-grasping in case of dropping objects under very large perturbations. Furthermore, the saliency map analysis method is used to describe weight distribution across various modalities during pinch grasping, confirming the effectiveness of our framework at leveraging multimodal information. The video is available online at: https://youtu.be/BlzxGgiKfck.

In human error-based learning, the size and direction of a scalar error (i.e., the "directed error") are used to update future actions. Modern deep reinforcement learning (RL) methods perform a similar operation but in terms of scalar rewards. Despite this similarity, the relationship between action updates of deep RL and human error-based learning has yet to be investigated. Here, we systematically compare the three major families of deep RL algorithms to human error-based learning. We show that all three deep RL approaches are qualitatively different from human error-based learning, as assessed by a mirror-reversal perturbation experiment. To bridge this gap, we developed an alternative deep RL algorithm inspired by human error-based learning, model-based deterministic policy gradients (MB-DPG). We showed that MB-DPG captures human error-based learning under mirror-reversal and rotational perturbations and that MB-DPG learns faster than canonical model-free algorithms on complex arm-based reaching tasks, while being more robust to (forward) model misspecification than model-based RL.

Traditional deep learning-based visual imitation learning techniques require a large amount of demonstration data for model training, and the pre-trained models are difficult to adapt to new scenarios. To address these limitations, we propose a unified framework using a novel progressive learning approach comprised of three phases: i) a coarse learning phase for concept representation, ii) a fine learning phase for action generation, and iii) an imaginary learning phase for domain adaptation. Overall, this approach leads to a one-shot domain-adaptive imitation learning framework. We use robotic pouring as an example task to evaluate its effectiveness. Our results show that the method has several advantages over contemporary end-to-end imitation learning approaches, including an improved success rate for task execution and more efficient training for deep imitation learning. In addition, the generalizability to new domains is improved, as demonstrated here with novel backgrounds, target containers, and granule combinations in the experiment. We believe that the proposed method is broadly applicable to various industrial or domestic applications that involve deep imitation learning for robotic manipulation, and where the target scenarios are diverse and human demonstration data is limited. For project video, please check our website:https://sites.google.com/view/imitation-learning-tase2022. Note to Practitioners—The motivation of this paper is to develop a progressive learning framework, which can be used for both service and industrial robots to learn from human demonstrations, and then transfer the learned skill to different scenarios with ease. We use the robotic pouring task as an example to demonstrate the effectiveness of our proposed method, since pouring is an essential skill for service robots to assist humans’ daily lives, and can benefit robot automation in wet-lab industries. The aim of this research is to enable robots to obtain visuomotor skills (such as the pouring skill), and accomplish the tasks with a high success rate using our proposed progressive learning method. We conducted experiments to show that the proposed method has good performance, high data efficiency and evident generalizability. This is significant for intelligent robots working in various practical applications.

Tactile pose estimation and tactile servoing are fundamental capabilities of robot touch. Reliable and precise pose estimation can be provided by applying deep learning models to high-resolution optical tactile sensors. Given the recent successes of Graph Neural Network (GNN) and the effectiveness of Voronoi features, we developed a Tactile Voronoi Graph Neural Network (Tac-VGNN) to achieve reliable pose-based tactile servoing relying on a biomimetic optical tactile sensor (TacTip). The GNN is well suited to modeling the distribution relationship between shear motions of the tactile markers, while the Voronoi diagram supplements this with area-based tactile features related to contact depth. The experiment results showed that the Tac-VGNN model can help enhance data interpretability during graph generation and model training efficiency significantly than CNN-based methods. It also improved pose estimation accuracy along vertical depth by 28.57% over vanilla GNN without Voronoi features and achieved better performance on the real surface following tasks with smoother robot control trajectories. For more project details, please view our website: https://sites.google.com/view/tac-vgnn/home

This paper presents a control scheme for force sensitive, gentle grasping with a Pisa/IIT anthropomorphic SoftHand equipped with a miniaturised version of the TacTip optical tactile sensor on all five fingertips. The tactile sensors provide high-resolution information about a grasp and how the fingers interact with held objects. We first describe a series of hardware developments for performing asynchronous sensor data acquisition and processing, resulting in a fast control loop sufficient for real-time grasp control. We then develop a novel grasp controller that uses tactile feedback from all five fingertip sensors simultaneously to gently and stably grasp 43 objects of varying geometry and stiffness, which is then applied to a human-to-robot handover task. These developments open the door to more advanced manipulation with underactuated hands via fast reflexive control using high-resolution tactile sensing.

Incipient slip is defined as the slippage of part, but not all, of the contact surface between a sensor and an object. Reliably detecting incipient slip in artificial tactile sensors would benefit autonomous robot handling capabilities by helping prevent object slippage during manipulation. Here, we present a biomimetic skin morphology based on the human fingerprint with application to marker-based tactile sensors such as the TacTip biomimetic optical tactile sensor. We modify the 3D-printed outer membrane of the TacTip to mimic glabrous skin morphology with the inclusion of external ridges (fingerprint) and internal markers (intermediate ridges), allowing localised shear deformation of the sensor's skin prior to the onset of gross slip. To validate the performance of this skin morphology, we train a random forest classifier (RFC) to identify incipient slip based on the extracted marker displacements from the sensor when it is compressed against an acrylic plate and moved laterally. The RFC model achieves 97.46% accuracy on incipient slip prediction, and is then validated on an unseen pouring task, in which gravity-induced incipient slip is detected on average within$418\pm 753\ \text{ms}$of its onset, and before gross slip in all trials. This accurate detection of incipient slip enables corrective actions prior to the onset of gross slip, a key capability in robotic manipulation and upper-limb prosthetics.

High-resolution tactile sensing can provide accurate information about local contact in contact-rich robotic tasks. However, the deployment of such tasks in unstructured environments remains under-investigated. To improve the robustness of tactile robot control in unstructured environments, we propose and study a new concept: tactile saliency for robot touch, inspired by the human touch attention mechanism from neuroscience and the visual saliency prediction problem from computer vision. In analogy to visual saliency, this concept involves identifying key information in tactile images captured by a tactile sensor. While visual saliency datasets are commonly annotated by humans, manually labelling tactile images is challenging due to their counterintuitive patterns. To address this challenge, we propose a novel approach comprised of three interrelated networks: 1) a Contact Depth Network (ConDepNet), which generates a contact depth map to localize deformation in a real tactile image that contains target and noise features; 2) a Tactile Saliency Network (TacSalNet), which predicts a tactile saliency map to describe the target areas for an input contact depth map; 3) and a Tactile Noise Generator (TacNGen), which generates noise features to train the TacSalNet. Experimental results in contact pose estimation and edge-following in the presence of distractors showcase the accurate prediction of target features from real tactile images. Overall, our tactile saliency prediction approach gives robust sim-to-real tactile control in environments with unknown distractors. Project page: https://sites.google.com/view/tactile-saliency/.

The sense of touch is fundamental to human dexterity. When mimicked in robotic touch, particularly by use of soft optical tactile sensors, it suffers from distortion due to motion-dependent shear. This complicates tactile tasks like shape reconstruction and exploration that require information about contact geometry. In this work, we pursue a semi-supervised approach to remove shear while preserving contact-only information. We validate our approach by showing a match between the model-generated unsheared images with their counterparts from vertically tapping onto the object. The model-generated unsheared images give faithful reconstruction of contact-geometry otherwise masked by shear, along with robust estimation of object pose then used for sliding exploration and full reconstruction of several planar shapes. We show that our semi-supervised approach achieves comparable performance to its fully supervised counterpart across all validation tasks with an order of magnitude less supervision. The semi-supervised method is thus more computational and labeled sample-efficient. We expect it will have broad applicability to wide range of complex tactile exploration and manipulation tasks performed via a shear-sensitive sense of touch.

In robots, nonprehensile manipulation operations such as pushing are a useful way of moving large, heavy, or unwieldy objects, moving multiple objects at once, or reducing uncertainty in the location or pose of objects. In this study, we propose a reactive and adaptive method for robotic pushing that uses rich feedback from a high-resolution optical tactile sensor to control push movements instead of relying on analytical or data-driven models of push interactions. Specifically, we use goal-driven tactile exploration to actively search for stable pushing configurations that cause the object to maintain its pose relative to the pusher while incrementally moving the pusher and object toward the target. We evaluate our method by pushing objects across planar and curved surfaces. For planar surfaces, we show that the method is accurate and robust to variations in initial contact position/angle, object shape, and start position; for curved surfaces, the performance is degraded slightly. An immediate consequence of our work is that it shows that explicit models of push interactions might be sufficient but are not necessary for this type of task. It also raises the interesting question of which aspects of the system should be modeled to achieve the best performance and generalization across a wide range of scenarios. Finally, it highlights the importance of testing on nonplanar surfaces and in other more complex environments when developing new methods for robotic pushing.

In this work, we report on how a sense of touch can be used to control an underactuated anthropomorphic robot hand, based on an integration that respects the hand’s mechanical functionality. Our focus is on integrating the sensorimotor control of the Pisa/IIT SoftHand, an anthropomorphic soft robot hand designed around the principle of adaptive synergies, with the BRL tactile fingertip (TacTip), a soft biomimetic optical tactile sensor. We consider: (i) closed-loop tactile control to establish a light contact on an unknown held object, based on the structural similarity of the tactile image; and (ii) controlling the estimated pose of a held object, using a convolutional neural network approach developed for other TacTip sensors. Accurate control was found for a range of hard and soft objects (to sub-millimetre accuracy and a few degrees). Overall, this gives a foundation to endow soft robotic hands with human-like touch, with implications for autonomous grasping, manipulation, human-robot interaction and prosthetics.

This work investigates uncertainty-aware deep learning (DL) in tactile robotics based on a general framework introduced recently for robot vision. For a test scenario, we consider optical tactile sensing in combination with DL to estimate the edge pose as a feedback signal to servo around various 2D test objects. We demonstrate that uncertainty-aware DL can improve the pose estimation over deterministic DL methods. The system estimates the uncertainty associated with each prediction, which is used along with temporal coherency to improve the predictions via a Kalman filter, and hence improve the tactile servo control. The robot is able to robustly follow all of the presented contour shapes to reduce not only the error by a factor of two but also smooth the trajectory from the undesired noisy behaviour caused by previous deterministic networks. In our view, as the field of tactile robotics matures in its use of DL, the estimation of uncertainty will become a key component in the control of physically interactive tasks in complex environments.

Tactile sensing is used by humans when grasping to prevent us dropping objects. One key facet of tactile sensing is slip detection, which allows a gripper to know when a grasp is failing and take action to prevent an object being dropped. This study demonstrates the slip detection capabilities of the recently developed Tactile Model O (T-MO) robotic hand by using support vector machines to detect slip and test multiple slip scenarios including responding to the onset of slip in real time with 11 different objects in various grasps. In this article, we demonstrate the benefits of slip detection in grasping by testing two real-world scenarios: adding weight to destabilize a grasp and using slip detection to lift up objects at the first attempt. The T-MO is able to detect when an object is slipping, react to stabilize the grasp, and be deployed in real-world scenarios. This shows the T-MO is a suitable platform for autonomous grasping by using reliable slip detection to ensure a stable grasp in unstructured environments.

Deep learning and reinforcement learning methods have been shown to enable learning of flexible and complex robot controllers. However, the reliance on large amounts of training data often requires data collection to be carried out in simulation, with a number of sim-to-real transfer methods being developed in recent years. In this paper, we study these techniques for tactile sensing using the TacTip optical tactile sensor, which consists of a deformable tip with a camera observing the positions of pins inside this tip. We designed a model for soft body simulation which was implemented using the Unity physics engine, and trained a neural network to predict the locations and angles of edges when in contact with the sensor. Using domain randomisation techniques for sim-to-real transfer, we show how this framework can be used to accurately predict edges with less than 1 mm prediction error in real-world testing, without any real-world data at all.

Developing artificial tactile sensing capabilities that rival human touch is a long-term goal in robotics and prosthetics. Gradually more elaborate biomimetic tactile sensors are being developed and applied to grasping and manipulation tasks to help achieve this goal. Here we present the neuroTac, a novel neuromorphic optical tactile sensor. The neuroTac combines the biomimetic hardware design from the TacTip sensor which mimicks the layered papillae structure of human glabrous skin, with an event-based camera (DAVIS240, iniVation) and algorithms which transduce contact information in the form of spike trains. The performance of the sensor is evaluated on a texture classification task, with four spike coding methods being implemented and compared: Intensive, Spatial, Temporal and Spatiotemporal. We found timing-based coding methods performed with the highest accuracy over both artificial and natural textures. The spike-based output of the neuroTac could enable the development of biomimetic tactile perception algorithms in robotics as well as non-invasive and invasive haptic feedback methods in prosthetics.

We present a modified TacTip biomimetic optical tactile sensor design which demonstrates the ability to induce and detect incipient slip, as confirmed by recording the movement of markers on the sensor's external surface. Incipient slip is defined as slippage of part, but not all, of the contact surface between the sensor and object. The addition of ridges - which mimic the friction ridges in the human fingertip - in a concentric ring pattern allowed for localised shear deformation to occur on the sensor surface for a significant duration prior to the onset of gross slip. By detecting incipient slip we were able to predict when several differently shaped objects were at risk of falling and prevent them from doing so. Detecting incipient slip is useful because a corrective action can be taken before slippage occurs across the entire contact area thus minimising the risk of objects been dropped.

Tactile sensing has been used for a variety of robotic exploration and manipulation tasks but a common constraint is a requirement for a large amount of training data. This paper addresses the issue of data-efficiency by proposing a novel method for online learning based on a Gaussian Process Latent Variable Model (GP-LVM), whereby the robot learns from tactile data whilst performing a contour following task thus enabling generalisation to a wide variety of tactile stimuli. The results show that contour following is successful with comparatively little data and is robust to novel stimuli. This work highlights that even with a simple learning architecture there are significant advantages to be gained in efficient and robust task performance by using latent variable models and online learning for tactile sensing tasks. This paves the way for a new generation of robust, fast, and data-efficient tactile systems.

Little research into tactile feet has been done for walking robots despite the benefits such feedback could give when walking on uneven terrain. This paper describes the development of a simple, robust and inexpensive tactile foot for legged robots based on a high-resolution biomimetic TacTip tactile sensor. Several design improvements were made to facilitate tactile sensing while walking, including the use of phosphorescent markers to remove the need for internal LED lighting. The usefulness of the foot is verified on a quadrupedal robot performing a beam walking task and it is found the sensor prevents the robot falling off the beam. Further, this capability also enables the robot to walk along the edge of a curved table. This tactile foot design can be easily modified for use with any legged robot, including much larger walking robots, enabling stable walking in challenging terrain.

Restoring tactile sensation is essential to enable in-hand manipulation and the smooth, natural control of upper-limb prosthetic devices. Here we present a platform to contribute to that long-term vision, combining an anthropomorphic robot hand (QB SoftHand) with a neuromorphic optical tactile sensor (neuroTac). Neuromorphic sensors aim to produce efficient, spike-based representations of information for bio-inspired processing. The development of this 5-fingered, sensorized hardware platform is validated with a customized mount allowing manual control of the hand. The platform is demonstrated to succesfully identify 4 objects from the YCB object set, and accurately discriminate between 4 directions of shear during stable grasps. This platform could lead to wide-ranging developments in the areas of haptics, prosthetics and telerobotics.

Manipulation tasks often require robots to be continuously in contact with an object. Therefore tactile perception systems need to handle continuous contact data. Shear deformation causes the tactile sensor to output path-dependent readings in contrast to discrete contact readings. As such, in some continuous-contact tasks, sliding can be regarded as a disturbance over the sensor signal. Here we present a shear-invariant perception method based on principal component analysis (PCA) which outputs the required information about the environment despite sliding motion. A compliant tactile sensor (the TacTip) is used to investigate continuous tactile contact. First, we evaluate the method offline using test data collected whilst the sensor slides over an edge. Then, the method is used within a contour-following task applied to 6 objects with varying curvatures; all contours are successfully traced. The method demonstrates generalisation capabilities and could underlie a more sophisticated controller for challenging manipulation or exploration tasks in unstructured environments.

Information technologies are permeating all aspects of manufacturing systems as well as other fields, expediting the generation of industrial big data. Traditionally, the devices collected the sensor data from various sources, and information fusion was then performed. This incurs higher burden of time and storage cost. Recently, more and more intelligent devices are equipped in the industrial environment. This provides more opportunities for better data collection and processing for industrial intelligence. Active perception technology, which performs control strategies on the data acquisition process, enables the devices to seamlessly integrate the perception and action to reach high-level goals rather than to accomplish low-level commands. It helps to select more useful information and may save the life of the sensors. However, there exist many unsolved challenging problems since the feedback is performed on complex processed sensory data, i.e., various extracted features.

Our human sense of touch enables us to manipulate our surroundings; therefore, complex robotic manipulation will require artificial tactile sensing. Typically tactile sensor arrays are used in robotics, implying that a straightforward way of interpreting multidimensional data is required. In this paper we present a simple visualisation approach based on applying principal component analysis (PCA) to systematically collected sets of tactile data. We apply the visualisation approach to 4 different types of tactile sensor, encompassing fingertips and vibrissal arrays. The results show that PCA can reveal structure and regularities in the tactile data, which also permits the use of simple classifiers such as k-NN to achieve good inference. Additionally, the Euclidean distance in principal component space gives a measure of sensitivity, which can aid visualisation and also be used to find regions in the tactile input space where the sensor is able to perceive with higher accuracy. We expect that these observations will generalise, and thus offer the potential for novel control methods based on touch.

There are a wide range of features that tactile contact provides, each with different aspects of information that can be used for object grasping, manipulation, and perception. In this paper inference of some key tactile features, tip displacement, contact location, shear direction and magnitude, is demonstrated by introducing a novel method of transducing a third dimension to the sensor data via Voronoi tessellation. The inferred features are displayed throughout the work in a new visualisation mode derived from the Voronoi tessellation; these visualisations create easier interpretation of data from an optical tactile sensor that measures local shear from displacement of internal pins (the TacTip). The output values of tip displacement and shear magnitude are calibrated to appropriate mechanical units and validate the direction of shear inferred from the sensor. We show that these methods can infer the direction of shear to ~2.3° without the need for training a classifier or regressor. The approach demonstrated here will increase the versatility and generality of the sensors and thus allow sensor to be used in more unstructured and unknown environments, as well as improve the use of these tactile sensors in more complex systems such as robot hands.

Here we propose and investigate a novel vibrissal tactile sensor - the Tac Whisker array - based on modifying a 3D-printed optical cutaneous (fingertip) tactile sensor - the TacTip. Two versions are considered: a static Tac Whisker array analogous to immotile tactile vibrissae (e.g. rodent microvib-rissae) and a dynamic Tac Whisker array analogous to motile tactile vibrissae (e.g. rodent macrovibrissae). Performance is assessed on an active object localization task. The whisking motion of the dynamic Tac Whisker leads to millimetre-scale location perception, whereas perception with the static Tac Whisker array is relatively poor when making dabbing contacts. The dynamic sensor output is dominated by a self-generated motion signal, which can be compensated by comparing to a reference signal. Overall, the Tac Whisker arrays give a new class of tactile whiskered robots that benefit from being relatively inexpensive and customizable. Furthermore, the biomimetic basis for the Tac Whiskers fits well with building an embodied model of the rodent sensory system for investigating animal perception.

Achieving object exploration with passive vision and active touch has been under investigation for thirty years. We build upon recent progress in biomimetic active touch that combines perception via Bayesian evidence accumulation with controlling the tactile sensor using perceived stimulus location. Here, passive vision is combined with active touch by providing a visual prior for each perceptual decision, with the precision of this prior setting the relative contribution of each modality. The performance is examined on an edge following task using a tactile fingertip (the TacTip) mounted on a robot arm. We find that the quality of exploration is a U-shaped function of the relative contribution of vision and touch; moreover, multi-modal performance is more robust, completing the contour when touch alone fails. The overall system has several parallels with biological theories of perception, and thus plausibly represents a robot model of visuo-tactile exploration in humans.

Tactile manipulation is the ability to control objects in real-time using the sense of touch. Here we examine tactile manipulation from the perspective of active touch with a biomimetic tactile sensor, which combines tactile perception with control of sensor location. Experiments are performed with the tactile fingertip mounted as end effector to a robot arm, to manipulate (roll) a cylinder in contact with the fingertip. Performance is validated with offline (cross-validation) and online (real-time operation) assessments. Location perception is finer than the sensor resolution, leading to superresolved tactile manipulation along a complex trajectory. However, the original methods were non-robust to large unknown disturbances of object location, necessitating modification of the perceptual process to diminish prior beliefs relative to past posterior beliefs. In consequence robust and accurate tactile manipulation was attained. In general, it appears there is a trade-off between the responsiveness to unknown change and manipulation accuracy, which must be set appropriately for each task.

Decision making under uncertainty is commonly modelled as a process of competitive stochastic evidence accumulation to threshold (the drift-diffusion model). However, it is unknown how animals learn these decision thresholds. We examine threshold learning by constructing a reward function that averages over many trials to Wald's cost function that defines decision optimality. These rewards are highly stochastic and hence challenging to optimize, which we address in two ways: first, a simple two-factor reward-modulated learning rule derived from Williams' REINFORCE method for neural networks; and second, Bayesian optimization of the reward function with a Gaussian process. Bayesian optimization converges in fewer trials than REINFORCE but is slower computationally with greater variance. The REINFORCE method is also a better model of acquisition behaviour in animals and a similar learning rule has been proposed for modelling basal ganglia function.

In this paper, we present an intrinsic motivation approach applied to haptics in robotics for tactile object exploration and recognition. Here, touch is used as the sensation process for contact detection, whilst proprioceptive information is used for the perception process. First, a probabilistic method is employed to reduce uncertainty present in tactile measurements. Second, the object exploration process is actively controlled by intelligently moving the robot hand towards interesting locations. The active behaviour performed with the robotic hand is achieved by an intrinsic motivation approach, which permitted to improve the accuracy for object recognition over the results obtained by a fixed sequence of exploration movements. The proposed method was validated in a simulated environment with a Monte Carlo method, whilst for the real environment a three-fingered robotic hand and various object shapes were employed. The results demonstrate that our method is robust and suitable for haptic perception in autonomous robotics.

Although superresolution has been studied to huge impact in visual imaging, it is relatively unexplored in tactile robotics. Here we demonstrate a novel optical sensor design (the TacTip) capable of achieving 40-fold localization superresolution to 0.1mm accuracy compared with a 4mm resolution between tactile elements. This superresolution is reached for localizing a 40mm diameter hemicylinder with a tactile finger pad also of 40mm diameter. Deformations of the sensor surface are measured as displacements of molded internal pins, with pin separation thus defining sensor resolution. Active Bayesian perception for classifying object location was used to ensure robust localization and hence the magnitude of the superresolution. These results are comparable with those for capacitive tactile sensors, which we interpret as originating from a convergence in the taxel-based design of the optical sensor and capacitive tactile sensors. The attained superresolution is comparable to the best perceptual hyperacuity in humans.

Embodied Choice considers action performance as a proper part of the decision making process rather than merely as a means to report the decision. The central statement of embodied choice is the existence of bidirectional influences between action and decisions. This implies that for a decision expressed by an action, the action dynamics and its constraints (e.g. current trajectory and kinematics) influence the decision making process. Here we use a perceptual decision making task to compare three types of model: a serial decision-then-action model, a parallel decision-and-action model, and an embodied choice model where the action feeds back into the decision making. The embodied model incorporates two key mechanisms that together are lacking in the other models: action preparation and commitment. First, action preparation strategies alleviate delays in enacting a choice but also modify decision termination. Second, action dynamics change the prospects and create a commitment effect to the initially preferred choice. Our results show that these two mechanisms make embodied choice models better suited to combine decision and action appropriately to achieve suitably fast and accurate responses, as usually required in ecologically valid situations. Moreover, embodied choice models with these mechanisms give a better account of trajectory tracking experiments during decision making. In conclusion, the embodied choice framework offers a combined theory of decision and action that gives a clear case that embodied phenomena such as the dynamics of actions can have a causal influence on central cognition.

Motivated by the impact of superresolution methods for imaging, we undertake a detailed and systematic analysis of localization acuity for a biomimetic fingertip and a flat region of tactile skin. We identify three key factors underlying superresolution that enable the perceptual acuity to surpass the sensor resolution: 1) the sensor is constructed with multiple overlapping, broad but sensitive receptive fields; 2) the tactile perception method interpolates between receptors (taxels) to attain subtaxel acuity; and 3) active perception ensures robustness to unknown initial contact location. All factors follow from active Bayesian perception applied to biomimetic tactile sensors with an elastomeric covering that spreads the contact over multiple taxels. In consequence, we attain extreme superresolution with a 35-fold improvement of localization acuity (0.12 mm) over sensor resolution (4 mm). We envisage that these principles will enable cheap high-acuity tactile sensors that are highly customizable to suit their robotic use. Practical applications encompass any scenario where an end-effector must be placed accurately via the sense of touch.

This work aims to augment the capacities for haptic perception in the iCub robot to generate a controller for surface exploration. The main task involves moving the hand over an irregular surface with uncertain slope, by concurrently regulating the pressure of the contact. Providing this ability will enable the autonomous extraction of important haptic features, such as texture and shape. We propose a hand controller whose operational space is defined over the surface of contact. The surface is estimated using a robust probabilistic estimator, which is then used for path planning. The motor commands are generated using a feedback controller, taking advantage of the kinematic information available by proprioception. Finally, the effectiveness of this controller is extended using a cerebellar-like adapter that generates reliable pressure tracking over the finger and results in a trajectory with less vulnerability to perturbations. The results of this work are consistent with insights about the role of the cerebellum on haptic perception in humans.

In this work, we present an active tactile perception approach for contour following based on a probabilistic framework. Tactile data were collected using a biomimetic fingertip sensor. We propose a control architecture that implements a perception-action cycle for the exploratory procedure, which allows the fingertip to react to tactile contact whilst regulating the applied contact force. In addition' the fingertip is actively repositioned to an optimal position to ensure accurate perception. The method is trained off-line and then the testing performed on-line based on contour following around several different test shapes. We then implement object recognition based on the extracted shapes. Our active approach is compared with a passive approach, demonstrating that active perception is necessary for successful contour following and hence shape recognition.

In this paper, we propose that active perception will help attain autonomous robotics in unstructured environments by giving robust perception. We test this claim with a biomimetic fingertip that senses surface texture under a range of contact depths. We compare the performance of passive Bayesian perception with a novel approach for active perception that includes a sensorimotor loop for controlling sensor position. Passive perception at a single depth gave poor results, with just 0.2mm uncertainty impairing performance. Extending passive perception over a range of depths gave non-robust performance. Only active perception could give robust, accurate performance, with the sensorimotor feedback compensating the position uncertainty. We expect that these results will extend to other stimuli, so that active perception will offer a general approach to robust perception in unstructured environments.

In a series of papers, we have formalized an active Bayesian perception approach for robotics based on recent progress in understanding animal perception. However, an issue for applied robot perception is how to tune this method to a task, using: (i) a belief threshold that adjusts the speed-accuracy tradeoff; and (ii) an active control strategy for relocating the sensor e.g. to a preset fixation point. Here we propose that these two variables should be learnt by reinforcement from a reward signal evaluating the decision outcome. We test this claim with a biomimetic fingertip that senses surface curvature under uncertainty about contact location. Appropriate formulation of the problem allows use of multi-armed bandit methods to optimize the threshold and fixation point of the active perception. In consequence, the system learns to balance speed versus accuracy and sets the fixation point to optimize both quantities. Although we consider one example in robot touch, we expect that the underlying principles have general applicability.

In this work, we apply active Bayesian perception to angle and position discrimination and extend the method to perform actions in a sensorimotor task using a biomimetic fingertip. The first part of this study tests active perception off-line with a large dataset of edge orientations and positions, using a Monte Carlo validation to ascertain the classification accuracy. We observe a significant improvement over passive methods that lack a sensorimotor loop for actively repositioning the sensor. The second part of this study then applies these findings about active perception to an example sensorimotor task in real-time. Using an appropriate online sensorimotor control architecture, the robot made decisions about what to do next and where to move next, which was applied to a contour-following task around several objects. The successful outcome of this simple but illustrative task demonstrates that active perception can be of practical benefit for tactile robotics.

Studies of decision making in animals suggest a neural mechanism of evidence accumulation for competing percepts according to Bayesian sequential analysis. This model of perception is embodied here in a biomimetic tactile sensing robot based on the rodent whisker system. We implement simultaneous perception of object shape and location using two psychological test paradigms: first, a free-response paradigm in which the agent decides when to respond, implemented with Bayesian sequential analysis; and second an interrogative paradigm in which the agent responds after a fixed interval, implemented with maximum likelihood estimation. A benefit of free-response Bayesian perception is that it allows tuning of reaction speed against accuracy. In addition, we find that large gains in decision performance are achieved with unforced responses that allow null decisions on ambiguous data. Therefore free-response Bayesian perception offers benefits for artificial systems that make them more animal-like in behavior.

Tactile sensing can be an important source of information for robots, and texture discrimination in particular is useful in object recognition and terrain identification. Whisker based tactile sensing has recently been shown to be a promising approach for mobile robots, using simple sensors and many classification approaches. However these approaches have often been tested in limited environments, and have not been compared against one another in a controlled way. A wide range of whisker-object contact poses are possible on a mobile robot, and the effect such contact variability has on sensing has not been properly investigated. We present a novel, carefully controlled study of simple surface texture classifiers on a large set of varied pose conditions that mimic those encountered by mobile robots. Namely, single brief whisker contacts with textured surfaces at a range of surface orientations and contact speeds. Results show that different classifiers are appropriate for different settings, with spectral template and feature based approaches performing best in surface texture, and contact speed estimation, respectively. The results may be used to inform selection of classifiers in tasks such as tactile SLAM.

Recent advances in modeling animal perception has motivated an approach of Bayesian perception applied to biomimetic robots. This study presents an initial application of Bayesian perception on an iCub fingertip sensor mounted on a dedicated positioning robot. We systematically probed the test system with five cylindrical stimuli offset by a range of positions relative to the fingertip. Testing the real-time speed and accuracy of shape and position discrimination, we achieved sub-millimeter accuracy with just a few taps. This result is apparently the first explicit demonstration of perceptual hyperacuity in robot touch, in that object positions are perceived more accurately than the taxel spacing. We also found substantial performance gains when the fingertip can reposition itself to avoid poor perceptual locations, which indicates that improved robot perception could mimic active perception in animals.

The basal ganglia are a subcortical group of interconnected nuclei involved in mediating action selection within cortex. A recent proposal is that this selection leads to optimal decision making over multiple alternatives because the basal ganglia anatomy maps onto a network implementation of an optimal statistical method for hypothesis testing, assuming that cortical activity encodes evidence for constrained gaussian-distributed alternatives. This letter demonstrates that this model of the basal ganglia extends naturally to encompass general Bayesian sequential analysis over arbitrary probability distributions, which raises the proposal to a practically realizable theory over generic perceptual hypotheses. We also show that the evidence in this model can represent either log likelihoods, log-likelihood ratios, or log odds, all leading proposals for the cortical processing of sensory data. For these reasons, we claim that the basal ganglia optimize decision making over general perceptual hypotheses represented in cortex. The relation of this theory to cortical encoding, cortico-basal ganglia anatomy, and reinforcement learning is discussed.

Many rodents use their whiskers to distinguish objects by surface texture. To examine possible mechanisms for this discrimination, data from an artificial whisker attached to a moving robot was used to test texture classification algorithms. This data was examined previously using a template-based classifier of the whisker vibration power spectrum. Motivated by a proposal about the neural computations underlying sensory decision making, we classified the raw whisker signal using the related `naive Bayes' method. The integration time window is important, with roughly 100ms of data required for good decisions and 500ms for the best decisions. For stereotyped motion, the classifier achieved hit rates of about 80% using a single (horizontal or vertical) stream of vibration data and 90% using both streams. Similar hit rates were achieved on natural data, apart from a single case in which the performance was only about 55%. Therefore this application of naive Bayes represents a biologically motivated algorithm that can perform well in a real-world robot task.