Combining physics-based simulation and machine learning for EIT-based tactile sensing

This work reports an innovative approach toward the development of tactile sensors based on electrical impedance tomography. Our approach couples the implementation of the forward model through a physics-based general purpose FEM software and an ANN model for the inverse problem solution. The ANN model is trained with an artificial dataset generated by using the forward model with different conductivity distributions. Simulations and tests on a real prototype shows promising performance.