Model-driven triboelectric sensors for multidimensional tactile perception

Tactile perception, as a key sensory function for environment awareness, has been the subject of extensive scientific research and technological application, producing artificial tactile sensors using piezoresistance, capacitance, magnetism to emerging triboelectricity. However, the decoupling of normal and shear loads remains elusive and faces a table-lookup challenge. Here, we report a tactile array sensor consisting of four triboelectric nanogenerator units with single-electrode-mode triboelectrification to realize the decoupling of normal and shear loads, where a unified theoretical model is developed based on a spinosum-hill small deformation hypothesis to address the table-lookup problem. The above decoupling of the normal and shear loads is further extended from a two-dimensional plane to a three-dimensional space based on the superposition principle of triboelectric outputs. Moreover, for a biological laboratory automation application, our sensor is installed on a robot manipulator to realize an automatic pipetting operation based on force feedback, solving the problem of droplet residue. We believe this study to be the very demonstration of model-driven tactile sensor for decoupling normal and shear loads, paving the way for the tactile perception of intelligent devices such as intelligent robotics.