Touch attention Bayesian models for robotic active haptic exploration of heterogeneous surfaces

  title={Touch attention Bayesian models for robotic active haptic exploration of heterogeneous surfaces},
  author={Ricardo Martins and Jo{\~a}o Filipe Ferreira and J. Dias},
  journal={2014 IEEE/RSJ International Conference on Intelligent Robots and Systems},
This work contributes to the development of active haptic exploration strategies of surfaces using robotic hands in environments with an unknown structure. The architecture of the proposed approach consists two main Bayesian models, implementing the touch attention mechanisms of the system. The model πper perceives and discriminates different categories of materials (haptic stimulus) integrating compliance and texture features extracted from haptic sensory data. The model πtar actively infers… 
Learning efficient haptic shape exploration with a rigid tactile sensor array
This work connects recent advances in recurrent models of visual attention with previous insights about the organisation of human haptic search behavior, exploratory procedures and haptic glances for a novel architecture that learns a generative model of haptic exploration in a simulated three-dimensional environment.
Combining Contact Forces and Geometry to Recognize Objects During Surface Haptic Exploration
This letter proposes an object recognition method, using a multivariate Gaussian–Bayesian classifier that collectively combines the haptic information, including friction coefficients and surface roughness, with the local geometry to recognize the object after surface haptic exploration.
Tactile-based active object discrimination and target object search in an unknown workspace
A tactile-based approach for estimating the center of mass of rigid objects and taking advantage of the prior knowledge obtained during the active touch learning, the robot took up to 15% fewer decision steps compared to the random method to achieve the same discrimination accuracy in active object discrimination task.
Active Tactile Transfer Learning for Object Discrimination in an Unstructured Environment Using Multimodal Robotic Skin
The experimental results show that using the proposed ATTL method, the robot successfully discriminated among new objects with 72% discrimination accuracy using only one training sample (on-shot-tactile-learning).
Active perception: Building objects' models using tactile exploration
The method uses Gaussian process (GPs) classification to efficiently sample the surface of the object in order to reconstruct its shape and outperforms random selection and previous work based on GP regression by sampling more points on and near-the-boundary of theobject.
Exploring Hardness and Geometry Information through Active Perception
A framework combining active perception and motion planning algorithm is proposed to get both hardness and geometry information of an object which also ensures working efficiency and Experimental results show that this framework has good performance and can explore global hardness and geography information efficiently.
Biomimetic Active Touch with Fingertips and Whiskers
  • N. Lepora
  • Biology
    IEEE Transactions on Haptics
  • 2016
Biomimetic active touch offers a common approach for biomimetic tactile sensors to accurately and robustly characterize and explore non-trivial, uncertain environments analogous to how animals perceive the natural world.
Planning and execution of groping behavior for contact sensor based manipulation in an unknown environment
This work proposes a method of sampling-based motion planning that enables the execution of the safe trial motion based on the criteria of feasibility and safety, and shows that a real robot plans and executes the manipulation with groping behavior in the occluded environment.
Tactile Object Recognition with Semi-Supervised Learning
It is proved that the proposed model of Ensemble Manifold Regularization outperforms the classic BoW framework and objects with similar features can be better classified.


Cerebellum-based adaptation for fine haptic control over the space of uncertain surfaces
This work proposes a hand controller whose operational space is defined over the surface of contact, whose effectiveness 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.
Tactile identification of objects using Bayesian exploration
The exploration algorithm was augmented with reinforcement learning whereby its internal representations of objects evolved according to its cumulative experience with them, allowing the algorithm to compensate for drift in the performance of the anthropomorphic robot hand and the ambient conditions of testing, improving accuracy while reducing the number of exploratory movements required to identify an object.
Surface material recognition through haptic exploration using an intelligent contact sensing finger
A novel haptic exploration strategy for recognizing the physical properties of unknown object surfaces using an intelligent finger that is capable of identifying the contact location, normal and tangential force, and the vibrations generated from the contact in real time is presented.
Feature Detection for Haptic Exploration with Robotic Fingers
The authors consider the detection of small surface features, such as cracks, bumps, and ridges, on the surface of an object during haptic exploration and dexterous manipulation and presents several algorithms for feature detection based on feature definitions.
Bayesian Exploration for Intelligent Identification of Textures
Performance of 99.6% in correctly discriminating pairs of similar textures was found to exceed human capabilities, and the method of Bayesian exploration developed and tested in this paper may generalize well to other cognitive problems.
Active contour following to explore object shape with robot touch
This work proposes 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.
A Control Framework for Tactile Servoing
This paper introduces a control framework to realize a whole set of tactile servoing tasks, i.e. control tasks that intend to realizes a specific tactile interaction pattern, by exploiting methods known from image processing.
Extracting data from human manipulation of objects towards improving autonomous robotic grasping
Tactile Feature Processing and Attentional Modulation in the Human Somatosensory System
The results of this study provided evidence for the direct engagement of feature-specific cortical areas in tactile perception and functions for feature processing and attentional modulation were added to the model by Dijkerman and de Haan.
Directions Toward Effective Utilization of Tactile Skin: A Review
The state of the art and the research issues in tactile sensing, with the emphasis on effective utilization of tactile sensors in robotic systems are surveyed, recognizing the fact that the system performance tends to depend on how its various components are put together.