Abstract

The dataset is related to grasp strength regulation in myocontrolled robotic hands, with a specific focus on the analysis of contact forces and adaptive control gain modulation during grasp execution. It includes force measurements and controller-related signals collected during human-in-the-loop experiments performed with an anthropomorphic robotic hand under different force control strategies. The dataset supports the quantitative evaluation and comparison of heuristic model-based, fuzzy-based, and neural network-based force controllers used to regulate grasp strength during tripod grasps. The data enable analysis of force tracking accuracy, interaction stability, and control smoothness across different grasp force levels. The data were acquired from experimental trials in which users controlled the robotic hand via surface electromyography (sEMG) signals and received vibrotactile feedback related to grasp force deviations. During the experiments, users were asked to track predefined target force levels while grasping rigid objects of different shapes. Contact forces at the robotic fingertips and the corresponding controller gain modulation signals were recorded during task execution to assess how different control strategies influence force overshoot, steady-state error, and responsiveness. The dataset enables statistical and qualitative analysis of force tracking performance, as presented in the associated publication: M. Sheikhsamad, R. Meattini, D. Chiaravalli, R. Suárez, J. Rosell, and G. Palli, “User-Tailored Fuzzy-Based Grasp Strength Regulation in Myocontrolled Robotic Hands,” IEEE 30th International Conference on Emerging Technologies and Factory Automation (ETFA), 2025