| In recent years,with the development of sensors and intelligent control technology,bionic robots that imitate human movements have been developed.However,the dexterity of these bionic robots still falls far short of human mobility.Capturing and monitoring human movement is the first step for bionic robots to learn human movements.At present,the most common way to capture human motion is optical camera,but this technology is limited by lighting conditions,and it is difficult to realize the real-time monitoring of subtle changes in the process of movement that are difficult to be detected by the naked eye.In addition,the current mechanical,electromagnetic,acoustic and other types of motion capture technology,have poor flexibility,long delay,poor resolution and other shortcomings,it is difficult to meet the needs of engineering.In addition,neither conventional inertial sensors nor commercial bending sensors are compatible with human skin.Therefore,the flexible strain sensor has become an effective player in motion capture.Because of its good tensile performance,the flexible tensile strain sensor can detect the joint flexion of the human body,but it is not sensitive to weak strain signals such as muscle contraction,so it cannot be detected,and there are problems that cannot be quantitatively analyzed.Aiming at the above problems,this thesis proposes to prepare a wearable carbon-based flexible bending sensor with high sensitivity,quantitative analysis and good performance.The main work is as follows:1.In order to prepare a flexible carbon-based bending sensor with high sensitivity and good stability,a carbon fiber beam with loose structure is used as the sensing material.In order to improve the sensitivity,the structural design is carried out based on the working principle.Finally,a cascaded carbon-based flexible sensor with three K-type sensing units is designed.Then,a packaging process to ensure the stability of the sensor is explored and selected,and can be arbitrarily combined with fabric to form a wearable bending sensor.Based on two working principles of contact area mechanism and quantum tunneling effect,a theoretical model is established and the relation between resistance and bending angle.The calibration experiment verifies the feasibility of the theoretical model and the consistency between the mathematical formula and the experimental data.Meanwhile,the error analysis of the theoretical Angle and the measurement angle is carried out.2.By means of flexible bending tester and various characterization methods,the performance of the bending sensor is investigated.It is found that the minimum resolution of the sensor can reach 0.5°.At the same time,Leica microscope was used to observe the elongation of the sensor during bending,and the calculated GF value reached 210,indicating the high sensitivity of the sensor.In addition,the response time of the sensor under the selfmade high frequency response device can reach 6.7ms,with fast response speed and good performance.Finally,the sensor can still maintain good performance under 51700 cycles,which proves that the sensor has good dynamic long-term durability.3.Due to the wide and sensitive working strain range of the sensor,it can not only monitor the human joint movement,but also monitor the subtle small strain of the human muscle surface.The sensor is applied to gesture recognition and grasp motion monitoring respectively.Firstly,a smart glove integrated with 14 sensors was prepared,which successfully recognized 10 digital gestures.Secondly,a set of upper limb movement acquisition system was prepared to monitor specific muscles and all joints in the grasping movement.Finally,visual processing of signal changes of upper limb muscles was realized,and the strength of muscle signals was illustrated by visual display effect.Based on a carbon-based flexible bending sensor,this thesis combines it with textile to make a wearable device.From the working mechanism to the formula derivation of the theoretical model,and then from the performance characterization to the practical application,the process of human movement is digitized through the bending sensor,which provides a reliable basis for the movement flexibility of bionic robots in the future. |
