Skin Simulation Models for Tactile Rendering

• Autores: Alvaro Gonzalo Perez Molero
• Directores de la Tesis: Miguel A. Otaduy Tristán (dir. tes.)
• Lectura: En la Universidad Rey Juan Carlos ( España ) en 2015
• Idioma: inglés
• Tribunal Calificador de la Tesis: Manuel Ferre Pérez (presid.) , Marcos Jose Garcia Lorenzo (secret.) , Eder Miguel Villalba (voc.) , Domenico Prattichizzo (voc.) , Maud Marchal (voc.)
• Enlaces
  • Tesis en acceso abierto en: hdl.handle.net
• Resumen

  • It is very common to discover around us objects that exhibit a hyperelastic mechanical behavior. Constitutive laws of hyperelastic materials relate stress and strain within a nonlinear regime, i.e., the deformation of this type of materials is not linear with respect to the force exerted on them. Examples of hyperelastic materials are: rubber, biological soft tissues and solid polymers, such as rubber-like materials. We are interested in modeling the mechanical behavior of the human skin tissue, in particular, the finger tissue.

    Constitutive models traditionally used to represent hyperelastic materials are too expensive computationally to be used in real-time physical simulations. In order to resolve this problem, we present a novel physically-based method that simulates accurately in real-time the extreme nonlinear elasticity of finger skin under frictional contact, showing rich and robust deformations in interactive scenarios. We use a nonlinear elasticity formulation using strain-limiting constraints, which is an efficient alternative to represent hyperelastic materials.

    To alleviate the computational cost of constrained dynamics, we propose a nonlinear constraints solver based on our nonlinear model with implicit integration. First, we present a projected Jacobi line-search solver that ensures a small cost per iteration and, next, we propose a novel nonlinear solver that approximates the actual nonlinear constraints in each iteration, in order to overcome the high nonlinear constraints present in our simulations, obtaining a solution to the problem faster than with standard solvers.

    We introduce the concept of soft finger tactile rendering for wearable cutaneous haptic devices. We propose an algorithm to command those cutaneous haptic devices that stimulate the skin through local contact surface modulation. To perform this, we solve an optimization problem which optimizes the degrees of freedom of each device to command, using our accurate nonlinear skin model, that allows us to simulate precisely the deformation of a real finger in contact with the cutaneous device.