IntelliMan_WP5_Grasping, Manipulation and Arm-Hand Coordination_T5.1_Data fusion and sensing technology_IEEESensorsLetters2026_Data

This dataset contains tactile signals acquired using a flexible piezoelectric sensing system during controlled experiments involving contact and slip events. The system consists of eight P(VDF-TrFE) piezoelectric sensing units fabricated on a flexible substrate and integrated into a biomimetic fingertip covered with a compliant silicone layer. Tactile signals were acquired from all eight sensing channels at a sampling frequency of 2 kHz. The dataset was collected using a three-axis Cartesian robot interacting with a 3D-printed rigid object. The experimental protocol involved three distinct tactile events: contact onset, slip onset, and contact release. Slippage was induced using a suspended weight, while a six-axis force/torque sensor provided reference measurements for identifying and labeling the tactile events. The experimental procedure was repeated 300 times, resulting in 900 labeled event recordings, with 300 observations per class. Additional experiments were conducted using three different surface textures (woven fabric, leather, and white paper), with 50 trials per texture, to investigate the robustness of the proposed recognition approaches under varying contact conditions. The dataset was used to investigate embedded machine learning approaches for real-time contact event recognition and slip onset detection. Two classification approaches were evaluated: a multilayer perceptron (MLP) trained on frequency-domain features and a one-dimensional convolutional neural network (1D-CNN) operating directly on tactile signals. The study also investigated the influence of temporal window size on classification accuracy and end-to-end latency, demonstrating the feasibility of deployment on resource-constrained embedded platforms. The dataset is associated with the following peer-reviewed publication, which provides a detailed description of the piezoelectric sensing system, experimental protocol, data acquisition methodology, and embedded machine learning framework: R. Khalifeh, M. Yaacoub, C. Gianoglio, M. Valle, and Y. Abbass, "Contact Events and Slip Onset Detection Based on a Piezoelectric Sensing System and Embedded Machine Learning," IEEE Sensors Letters, vol. 10, no. 8, Art. no. 5504804, 2026. DOI: https://doi.org/10.1109/LSENS.2026.3714000 Users of this dataset are kindly requested to cite the above publication when using the data in their research.