Vision-Based Tactile Sensing and Shape Estimation Using a Fluid-Type Touchpad

In this paper, we propose a new method to estimate the shape and irregularity of objects by a vision-based tactile sensor, which consists of a CCD camera, LED lights, transparent acrylic plate, and a touchpad which consists of an elastic membrane filled with translucent red water. Intensities of red, green and blue bands of the traveling light in the touchpad are analyzed in this study to estimate the shape/irregularity of the object. The LED light traveling in the touchpad is scattered and absorbed by the red pigment in the fluid. The depth of the touchpad is estimated by using the intensity of the light obtained from the red-green-blue (RGB) values of the image, in consideration of the scattering and reflection effects. The reflection coefficient that depends on the shape of the membrane, was decoupled in the proposed formulation. The intensity of the traveling light is represented with the geometrical parameters of the touchpad surface. In order to reduce the approximation error caused by unmodeled factors, we compensate the error by using a function of the deformation of the membrane. The validation of the proposed method is confirmed through experimental results.