Research On Multimodal Information Measurement Method Of Tactile Perception Evoked By Skin Friction

The development of artificial intelligence has produced tremendous value for society,and tactile sensation has hindered its development in artificial intelligence due to its unclear sensing mechanism.The tactile perception process includes the four elements of touch object attributes,friction process,brain response,and brain cognition,and involves various fields such as tribology,psychology,and neuroscience.Most of the existing literature studies single change conditions and individual modes such as touch speed,touch samples,three-dimensional tactile force,vibration,EEG,etc.Studying tactile perception in this way is inevitably limited to a certain degree.In order to describe the tactile perception process systematically and comprehensively,we proposes a multi-modal haptic information measurement method and then designs the hardware and software of the haptic device to realize the functions we need,and then design a tactile multi-modal perception experiment with roughness as the variable based on the haptic device,neuroelectrophysiological equipment and three-dimensional surface topography instrument.Finally,the characteristics of tactile modal and the correlation between the modalities related to roughness were studied and feasibility of multi-modal haptic measurement methods were analyzed.The research is carried out through the following parts:1.Analyze the existing research methods of tactile perception,and compare the advantages and disadvantages of the haptic device.Based on the above work,we propose a multi-modal haptic information measurement method,and design the software and hardware of the haptic device through a modular approach to enable it to simultaneously collect touch pressure,friction and vibration.The collected signal can be observed and saved in the haptic software.In addition,researchers can change the touch speed,touch circles,touch samples to flexible design haptic experiments.Multimodal haptic exploration would carried out well by our haptic device,threedimensional surface profilometer,EEG,f NIRS and other electrophysiological signal acquisition equipment.2.Haptic multimodal acquisition experiment in this thesis was designed with roughness as the variable.It is divided into three parts: object surface characteristic measurement experiment,psychophysics experiment,friction mechanics and EEG signal acquisition experiment.The collected modal quantities are processed and then features of modal quantities are extracted.Ra,Rz,Rsm,particle size and mesh are chosen as the surface features of the object.Pressure,friction,acceleration power spectrum center of gravity and its average power are chosen as frictional mechanics quantities.The ratio of alpha-wave and beta-wave energy of each electrode are chosen as EEG signal feature.Roughness score is counted in the subjective evaluation.Check whether each modal feature value has a significant difference between samples through sample paired t test,and calculate the Pearson correlation coefficient between significant modal feature values,and the following results are showed: 1)3000 mesh and 600 mesh roughness scores,average pressure,average friction,acceleration power spectral centroid and alpha-wave and beta-wave energy ratios in part of the electrodes are significantly different,and the differences between other samples are not necessarily significant among individuals.2)3000-mesh and 600-mesh roughness scores,frictional forces,and the alpha-wave,beta-wave energy ratios of FCz,Cz,and POz electrode have a moderate or little correlation.This result proves that the multimodal measurement method of tactile perception has certain feasibility.