A novel haptic deformation modelling method using spherical harmonic representation and radial basis function neural network

作者：Fang Yanhong; Wu Bin^*; Yang Zhengyi

来源：International Journal of Digital Content Technology and Its Applications, 2011, 5(10): 105-111.

DOI：10.4156/jdcta.vol5.issue10.13

• Applied forces
• Computational time
• Deformation models
• Error measures
• Modelling method
• Multi-resolutions
• Neural network model
• Object surface
• Radial basis function neural networks
• Radial basis functions
• RBF Neural Network
• Real-time application
• Real-time deformation
• Rotational invariances
• Simulation result
• Soft objects
• Spherical harmonics
• Young's Modulus

In the haptic modelling of deformable soft objects in virtual reality, due to the complicated relationship between the deformation and the applied force, it is a challenging task to achieve the balance between accuracy and efficiency. In this paper, a novel approach based on spherical harmonic (SH) representation and Radial Basis Function neural network(RBF-NN) to simulating the virtual deformation of soft objects was proposed. The object surface was represented by SH, which features of orthonormality, rotational invariance, and multi-resolution. Given the density, Young's modulus, Poisson ratio of the object and interactive position of the applied force, the haptic deformation model was estimated by a Radial Basis RBF) based neural network model, which was trained with Least Error Measure method. The simulation results indicated that the haptic deformation model could be accurately calculated using the proposed approach and the computational time was acceptable for real-time applications.

• 出版日期2011
• 单位西南大学; university of Queensland; The University of Queensland