Surface material recognition through haptic exploration using an intelligent contact sensing finger

Abstract

Object surface properties are among the most important information which a robot requires in order to effectively interact with an unknown environment. This paper presents a novel haptic exploration strategy for recognizing the physical properties of unknown object surfaces using an intelligent finger. This developed intelligent finger is capable of identifying the contact location, normal and tangential force, and the vibrations generated from the contact in real time. In the proposed strategy, this finger gently slides along the surface with a short stroke while increasing and decreasing the sliding velocity. By applying a dynamic friction model to describe this contact, rich and accurate surface physical properties can be identified within this stroke. This allows different surface materials to be easily distinguished even if when they have very similar texture. Several supervised learning algorithms have been applied and compared for surface recognition based on the obtained surface properties. It has been found that the naïve Bayes classifier is superior to radial basis function network and k-NN method, achieving an overall classification accuracy of 88.5% for distinguishing twelve different surface materials.

Global estimation of an object’s pose using tactile sensing

“…Equation 6 and Figure 12 illustrate this scheme where, using 6 axis force-torque sensing under a parametrisable convex shape S -a semi-ellipsoid in this case -one can solve the system of equations consisting of the surface equation and the force F = [F x , F y , F z ] and moment M = [m x , m y , m z ] balance, yielding a unique solution for the contact location p c and a torque τ around the contact normal n. From here, it is trivial to decompose the total measured force into its normal and tangential components. This approach has been previously validated by the authors in Liu et al [48] showing an accuracy of 0.224 mm in contact location, corresponding to an error of 1.5 • in contact normal. This sensing strategy has found several successful applications, ranging from slip detection [49] to surface following [50].…”

Re-using prior tactile experience by robotic hands to discriminate in-hand objects via texture properties

“…Liu et al employed an intelligent contact sensing finger to classify surface materials with Naive Bayes classifier [3]. Hu et al used Support Vector Machine (SVM) to classify five different fabrics by sliding a finger-shaped sensor over the surfaces [4].…”

Cerebellum-based adaptation for fine haptic control over the space of uncertain surfaces

“…(8) whereR [s,3] is the third row of the current rotation matrixR s .L e t J [3] (q) be deÞned as the third row of the kinematic Jacobian, parameterized by the current joints measurementq. Thus, the measurement of the vertical hand displacement can be computed bŷ…”

“…An important task in Þne exploration is to place the hand over an unknown surface and slide it while regulating the pressure of the contact. Tactile arrays and soft Þngers have been commonly used to execute this exploratory procedure, in order to achieve surface recognition [8] or reconstruction [12]. SpeciÞcally, the work in [11] relies on a soft Þngertip that allows measuring the contact force from any direction.…”