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

Abstract

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 deÞned 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 cerebellarlike adapter that generates reliable pressure tracking over the Þnger 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.

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

“…A second class of works integrates sensing, perception and local exploration mechanisms similar to the previous works, however they expand the exploration strategy to large and heterogeneous surfaces in the haptic features domain under analysis. The perceptual haptic map of the surface can be constructed following different strategies: the global exploration path is fixed and defined a-priori [18] [19] [20], the exploration is performed actively showing a active behaviour [21], [22]. This work contributes to this class of approaches by proposing a formulation of Bayesian models implementing touch attention mechanisms involved in the active haptic exploration of unknown surfaces.…”

“…This metric determines the total divergence between the exploration path executed by the robotic platform V and the benchmark path B defined by an external operator. The definition of this metric follows the same principles of analogous metrics proposed by [22] and [20].…”