Bioinspired Co‐Design of Tactile Sensor and Deep Learning Algorithm for Human–Robot Interaction

摘要

Human–Robot Interaction In article number 2200050, Geng Yang and co-workers implement the co-design of deep learning algorithms and the tactile sensor. By utilizing deep neural networks (DNNs), the tactile sensor can effectively detect the location and magnitude of external force and recognize different touch modalities. Besides, a novel data augmentation method is developed based on the sensor’s rotation symmetry structure, which enhances the DNNs’ generalization performance.