抄録
The perception of surface roughness during active tactile exploration is strongly influenced by friction-induced vibrations generated at the contact interface. In friction tests using sinusoidal reciprocating motion that reproduces human tactile exploratory movements, acceleration and deceleration during sliding hinder accurate evaluation of the vibration components attributable to surface roughness when conventional time-domain frequency analysis is employed. This study proposes a position-equidistant spectral analysis method that resamples friction force signals at spatially equidistant intervals based on positional data. First, the validity of the proposed method is verified using artificial roughness samples with known protrusion spacing, demonstrating that distinct spatial frequency peaks corresponding to protrusion spacing can be extracted. Then, the method is applied to two groups of practical materials: Group 1, comprising bovine leathers, synthetic leathers, coating films, and polymers, and Group 2, comprising knit fabrics with different structural characteristics. The results demonstrate that the friction amplitude within specific spatial frequency bands, corresponding to the fingerprint ridge spacing of the tactile contactor, exhibits a clear relationship with the perception of roughness obtained from sensory evaluations. These findings indicate that spatial frequency–based evaluations of friction-induced vibrations are effective for assessing roughness-related tactile perception under reciprocating friction conditions.
