IntelliMan_WP5_Grasping, Manipulation and Arm-Hand Coordination_T5.1_Data fusion and sensing technology_IEEESENSORSLETTERS2023_Data

The dataset has been collected on three touch modalities, i.e. slide a finger, brush a paintbrush, and rolling a washer. The dataset is made of 840 tensor data, 280 per class. Seventy subjects performed each touch action two times on a 4 × 4 piezoelectric sensors matrix in two directions, i.e. horizontal and vertical starting from a random position. Each touch lasted 10 seconds and signals have been sampled at 3kHz. Each datum is a 3-mode tensor where the first two modes represent the 4 × 4 sensors geometry (i.e. a total of 16 channels) and the third dimension contains the time samples (10s × 3K Samples/s = 30000 samples per each channel)