Haptic SLAM: An Ideal Observer Model for Bayesian Inference of Object Shape and Hand Pose from Contact Dynamics
Dynamic tactile exploration enables humans to seamlessly estimate the shape of objects and distinguish them from one another in the complete absence of visual information. Such a blind tactile exploration allows integrating information of the hand pose an
- PDF / 4,481,640 Bytes
- 12 Pages / 439.37 x 666.142 pts Page_size
- 1 Downloads / 178 Views
Department of Computing, Imperial College London, London, UK {feryal,a.faisal}@imperial.ac.uk Department of Bioengineering, Imperial College London, London, UK MRC Clinical Sciences Centre, Imperial College London, London, UK
Abstract. Dynamic tactile exploration enables humans to seamlessly estimate the shape of objects and distinguish them from one another in the complete absence of visual information. Such a blind tactile exploration allows integrating information of the hand pose and contacts on the skin to form a coherent representation of the object shape. A principled way to understand the underlying neural computations of human haptic perception is through normative modelling. We propose a Bayesian perceptual model for recursive integration of noisy proprioceptive hand pose with noisy skin–object contacts. The model simultaneously forms an optimal estimate of the true hand pose and a representation of the explored shape in an object–centred coordinate system. A classification algorithm can, thus, be applied in order to distinguish among different objects solely based on the similarity of their representations. This enables the comparison, in real–time, of the shape of an object identified by human subjects with the shape of the same object predicted by our model using motion capture data. Therefore, our work provides a framework for a principled study of human haptic exploration of complex objects. Keywords: Tactile sensing · Haptic exploration · Internal representation · Ideal observer · Object recognition · Normative model
1
Introduction
Enabled by the remarkable dexterity of our hands and the multi–modal tactile sensors of our skin, 3–dimensional objects can be effortlessly detected and categorised from among thousands of possibilities. This is possible even in the absence of visual feedback and despite the presence of variability in our sensory and motor pathways introducing uncertainty in the true pose of our hands and the precise location of contacts on our skin [5,12]. This uncertainty complicates the localisation of tactile contact sensations in forming the internal representation of object shape and its location within the environment. Consider using c Springer International Publishing Switzerland 2016 F. Bello et al. (Eds.): EuroHaptics 2016, Part I, LNCS 9774, pp. 146–157, 2016. DOI: 10.1007/978-3-319-42321-0 14
Haptic SLAM
147
a single finger to haptically explore an unknown object. The finger moves in space and we register contact locations on our skin. From the information of our finger’s pose and the location of contacts, we can estimate where in space with respect to our hand the surface of the object lies. This is complicated as due to the presence of proprioceptive noise, the precise location of the finger within the space is not known. Also, the exact location where the contact between the object and the skin has occurred is ambiguous. Tactile exploration involves repeatedly exploring the space and piecing together out of the multiple contacts a tactile representation of how the object ma
Data Loading...
