Bayesian Exploration for Intelligent Identification of Textures

By Jeremy A. Fishel, Gerald E. Loeb. Front. Neurorobot., 2012

380 citations recorded by the source corpus.

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.

Subfields: Biomedical Engineering, Cognitive Neuroscience

View Paper: https://www.semanticscholar.org/paper/e3e9b9527f141da240e21333a65526565e049b3f

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Biomedical Engineering

In Cognitive Neuroscience

Citations in the corpus

Learning tactile skills through curious exploration (2012, background): “Whereas Fishel and Loeb (2012) learn to select pre-scripted exploratory movements, our algorithm can still reﬁne the previously-learned exploratory movements during the learning of the supervised texture classiﬁcation task.”
Robotic Tactile Perception of Object Properties: A Review (2017, methodology): “Using a BioTac sensor in [67], Bayesian exploration is proposed to discriminate textures, which selects the optimal movements adaptively based on”
A Review of Tactile Information: Perception and Action Through Touch (2020, background): “In-spiredbytheseﬁndings,manytactile-basedexplorationmethods havebeendevelopedforrobotstoestimateimportantparameters includingcontact-levelinformationsuchaslocalgeometry[71] andmaterialproperties[106],aswellasobject-levelinformation suchasshape[169],[170]andmassinformation[134].”
Robotics and Autonomous Systems (2014, neutral): “Although a system capable of sensing different types of information in parallel is generally desired [51,52,67,145], this does not necessarily lead to significant improvements in performance.”
Tactile sensing in dexterous robot hands - Review (2015, methodology): “slippage [41], to detect a contact of a grasped object with an environment [38] or to recognize a texture of a surface [22], then the response time of a sensor becomes crucial.”
Electronic skins and machine learning for intelligent soft robots (2020, background): “Hence, there have been studies on acquiring these specialized closed-loop policies based on intrinsic rewards such as curiosity (99) or extrinsic rewards such as texture discrimination ability (100).”
Interactive Perception: Leveraging Action in Perception and Perception in Action (2016, background): “More recent examples include [23] to learn haptic object representations, [34]–[36] for object detection and pose estimation, [37]–[39] for reconstructing the shape of objects or the environment, and [40]–[42] for texture classification or description.”