From Pixels to Percepts: Highly Robust Edge Perception and Contour Following Using Deep Learning and an Optical Biomimetic Tactile Sensor

Abstract

Deep learning has the potential to have the impact on robot touch that it has had on robot vision. Optical tactile sensors act as a bridge between the subjects by allowing techniques from vision to be applied to touch. In this paper, we apply deep learning to an optical biomimetic tactile sensor, the TacTip, which images an array of papillae (pins) inside its sensing surface analogous to structures within human skin. Our main result is that the application of a deep CNN can give reliable edge perception and thus a robust policy for planning contact points to move around object contours. Robustness is demonstrated over several irregular and compliant objects with both tapping and continuous sliding, using a model trained only by tapping onto a disk. These results relied on using techniques to encourage generalization to tasks beyond which the model was trained. We expect this is a generic problem in practical applications of tactile sensing that deep learning will solve.

Tactile Image Sensors Employing Camera: A Review

“…A dotted pattern is printed on the inside of the touchpad surface. Recently, it has become possible for soft materials to be processed by 3D printers, and the elastic part with markers can be directly fabricated through 3D printing [27,35,67,68,69,70,71,72].…”

Tactile Model O: Fabrication and Testing of a 3D-Printed, Three-Fingered Tactile Robot Hand

“…Recent research has shown that this second technique gives improved results in a contour following tasks, particularly when concerned with robustness in an online setting. 40 Thus, in this article, the main approach will be to use raw images and neural networks; the hand has, however, been designed to accommodate both approaches.…”

“…To estimate the appropriate frame, we calculate the absolute pixel difference for each frame in a tactile video when compared with the first frame in the same video. 40 This gives a basic measure of sensor deformity.…”

Deep Reinforcement Learning for Tactile Robotics: Learning to Type on a Braille Keyboard

“…Following recent work with this tactile sensor [22], the preprocessed image captured by the sensor is directly passed into the machine learning algorithms. This removes the need for pin detection and tracking algorithms that were necessary in previous work, and enables the advantages of convolutional neural networks to be applied to tactile data.…”