Gesture Recognition System With Two-Dimensional Material Composite-Based Flexible Pressure Sensors

In recent years,two-dimensional materials have been widely used in the wearable field due to their excellent mechanical and electrical properties.Compared with traditional silicon crystals,typical two-dimensional materials such as graphene and transition metal dichalcogenides have the characteristics of high carrier mobility and higher mechanical strength,and can be used as flexible pressure sensing materials.The pressure sensor based on the piezoresistive effect can be used to collect the pressure change signal generated by the movement of various parts of the human body,and complete the recognition of the human body posture by analyzing the change law of the piezoresistive signal.In this paper,a graphene-based two-dimensional material composite is used to prepare a highly compressible sensor with piezoresistive effect.The pressure sensor array is placed at the finger joints,the pressure sensor array signals are collected,and the signals are processed and analyzed at the sensor array end to complete the intelligent recognition of gestures combining with the neural network algorithm.The specific research results are as follows:?1?Graphene and vanadium diselenide composites are used as pressure sensitive materials,flexible copolyester is used as the three-dimensional framework,and highly compressible flexible pressure sensors are integrated.The materials and the microstructure of the sensors are characterized by advanced testing methods.The sensor has good linearity in the range of low pressure?<10 k Pa?and medium and high pressure??10 k Pa?.The performance of the composite sensor with different ratios of graphene and vanadium diselenide was tested,and it was found that when the concentration of graphene and vanadium diselenide were both 0.6 g/L,the maximum relative resistance change could reach 81.6%,the maximum The deformation is 85%,the response time is 300 ms,and the sensitivity reaches 4 k Pa^-1.?2?The above-mentioned two-dimensional material composite pressure sensor is assembled at the finger's pressure sensitive point,namely the finger joint,to collect the pressure change signal generated when the finger is bent.The sensor array is packaged on a glove,and the collected array voltage-transformation signal is transmitted to the signal processing end through a voltage follower circuit.The signal processing end is divided into static gesture recognition and dynamic object grasping recognition.Static gesture recognition uses a low-power CMOS chip combining with a binary neural network algorithm,and has an accuracy of 98%for Arabic numerals and dumb gesture recognition.Dynamic object recognition uses a general-purpose CPU combining with the BP networks processing timing signals to complete object grabbing and recognition with an accuracy rate of 89%by extracting and analyzing the continuous changes of the pressure signal when grabbing conventional objects such as water bottles and mobile phones.In summary,this paper uses two-dimensional materials represented by graphene to prepare a flexible pressure sensor,and completes the integration of the smart gesture recognition system by combining with advanced artificial intelligence algorithms.This work promotes the development of two-dimensional materials based flexible smart sensors in the wearable field.