Deep Gaussian Processes for Angle and Position Discrimination in Active Touch Sensing

Abstract

Active touch sensing can benefit from the representation of uncertainty in order to guide sensing movements and to drive sensing strategies that operate to reduce uncertainty with respect to the task at hand. Here we explore learning approaches that can acquire task knowledge quickly and with relatively small datasets and with the potential to be exploited for active sensing in robots and as models of biological sensory systems. Specifically, we explore the utility of deep (hierarchical) Gaussian Process models (Deep GPs) that have shown promise as models of episodic memory processes due to their low-dimensionality (compactness), generative capability, and ability to explicitly represent uncertainty. Using data obtained in a robotic active touch task (contour following), we show that both single-layer and Deep GP models are capable of providing robust function approximations from tactile data to angle and sensor position, with Deep GPs showing some advantages in terms of accuracy and uncertainty quantification in angle discrimination.

Simple Synthetic Memories of Robotic Touch

“…It has been previously demonstrated that linear dimensionality reduction of systematically collected tactile data involving orientation and sensor position produces a structured manifold. The generated manifold reflects regularities from the observational space with the potential to support accurate perception of magnitudes for sensorimotor control [1]; specifically in the discrimination of angle and position of the sensor with respect to the edge of an object using non-parametric models in supervised learning [14]. Alternatively, non-linear dimensionality reduction has been studied in the context of learning a manifold to relate tactile data and actions for object recognition [17].…”