Multi-scale Flexible Touch-pressure Sensor Based On 2D/3D Graphene

Bionic electronic skin simulates and interacts with the external environment by combining micro-electromechanical,materials science and sensor technologies.Therefore,bionic electronic skin is an important medium for wearable devices and intelligent robots to sense the external environment.As an important part of bionic electronic skin,the touch-pressure sensor can achieve multi-scale measurement of large pressure while capturing small pressure,which reduces the perception process of human skin on mechanics.Therefore,the touch-pressure sensor has great development prospects in the fields of wearable devices,medical devices and intelligent robots.The human skin has a micro-pressure high-precision?tactile?for pressure sensing and a large pressure?pressure?sensing function.Because the single pressure sensor is inconsistent in measuring range and precision,it is difficult to meet the human body's external force demand.Therefore,the research on biomimetic pressure sensors at home and abroad is divided into two categories according to the measurement range:one is high sensitivity touch.Sensors,which have high sensitivity,but the measurement range is usually small.The other type is a pressure sensor with a large measuring range.Although the sensor can measure a large pressure range,it is usually rigid and has low measurement accuracy in a small pressure range.Most of the bionic pressure sensors studied can only be regarded as a single sensor of tactile or embossing,and it is difficult for these sensors to simultaneously simulate the impression of pressure and touch.In view of the above problems,this paper proposes a multi-scale flexible touch-pressure sensor based on 2D/3D graphene.In this paper,a flexible touch-pressure sensor based on 2D/3D graphene is proposed by studying the principle of human skin sensing external force,the working principle of graphene under 2D and 3D structure and the mutual constraint relationship.By investigating the physical parameters of the human skin's perceived external force,the structure,size and key parameters of the sensor were studied.The contact-pressure sensor unit was realized through the material selection and preparation process of the sensor.Then the micro-pressure-pressure experimental measurement platform was built.And through the experimental analysis to study the performance of the sensor.The experimental results show that the measurement range of this sensor is in the range of 0²N and 2⁴0N.For tactile sensors,the sensitivity can reach 472.2?/kPa,the force resolution can reach 0.01N,the response time is about 40ms;the sensitivity of the pressure sensor can reach5.05k?/kPa,the force resolution can reach 0.5N,and the response time is about 30ms.The touch-pressure sensor proposed in this paper has multiple sensitive units and is a single input-multiple output system.When the sensor is calibrated,the least squares method is usually used to determine the corresponding functional relationship between the sensor input and the output.This method has better performance in dealing with linear systems.The actual results of the sensors in this paper are nonlinear,and the least squares method is more complicated to deal with multi-input-multiple output systems.Therefore,this paper proposes a calibration method based on SVM for bionic touch-pressure sensor.Through the error analysis of the calibration results,it is found that when the SVM algorithm is used to calibrate the sensor,the mean square error of the tactile sensor is5.370327×10^-5,and the pressure sensor is 1.278634×10^-2.The calibration result is two orders of magnitude smaller than the mean square error of the traditional least squares method.The bionic touch-pressure sensor proposed in this paper is flexible,and the measurement range spans 3 scales,which can simulate the function of human skin to sense external force.The SVM-based sensor calibration method can improve the calibration efficiency and accuracy of single-input-multi-output systems.This method also proposes a new processing method for other multi-input-multiple output sensor systems.The research results have important application prospects in mechanical arm,biomimetic robot and sensor calibration.