Parameter estimation and object gripping based on fingertip force/torque sensors

Object parameters, like the mass and barycenter, are significant for the stability of object gripping. This study proposes a novel method to estimate the mass and barycenter coordinate of the task object based on the fingertip force/torque sensor. The equilibrium conditions of force and torque are deduced, and the estimation models are built for mass and barycenter coordinate. In particular, the principal component analysis (PCA) method is used to reduce inter-dimensional coupling and extract principal components for the multi-dimensional and multi-redundant signals of the fingertip force/torque sensor. Furthermore, an object parameter-based method is proposed for the task object gripping. Finally, we conduct a set of experiments to validate the effectiveness and generalization of the proposed method. The experiment results show that the parameter estimation method is effective to estimate the mass and barycenter coordinate, and the object parameter-based method can be generalized to grip irregular objects with varying gripping points.