Research On Humanoid Head Robot And Expression Realization Method

With the continuous development of intelligent robot technology,humanoid head robots that can realize functions such as facial expressions have gradually become a research hotspot.At present,humanoid robots have many driving points,the degrees of freedom that can be achieved are limited,and most of them are driven by electric motors.Although they have the advantages of large torque and precise motion,they also have the disadvantages of taking up a lot of space and being heavy.In summary,in view of the problems of human head robots,this paper mainly completes the following research contents:Based on a detailed review of the research status of humanoid head robots at home and abroad,based on the analysis of facial muscles and the facial action coding system(FACS),the muscle groups that control facial organ movements were analyzed.For six facial expressions,analyze their facial morphological characteristics and action descriptions,set 7 control points,and use different motion conversion methods to achieve dynamic conversion between expressions.The mechanical structure system and motion control system of the humanoid head robot are designed.The mechanical structure includes neck mechanism,eye mechanism,jaw mechanism,etc.The left and right movements of two humanoid eyeballs in the eye mechanism can be controlled separately,which increases the eye movement flexibility.According to the actual situation of the humanoid head robot,this paper adopts multiple driving methods such as steering gear,shape memory alloy(SMA),the ion-exchange polymer metal composite material(IPMC)to drive and design modularly.The neck structure is driven by the steering gear,the eyebrows are driven by the IPMC,and the eye mechanism,the jaw mechanism and the facial skin are driven by the SMA driver.In addition,the overall control scheme of the humanoid head robot was designed,using Arduino as the controller,and using Matlab/Simulink and Arduino IDE as the upper computer to control each mechanism.Finite element analysis(FEA)is used to simulate facial expressions.First,use surface modeling to create a three-dimensional model of the facial skin in Solidworks,and then import it into Hypermesh to perform geometric pre-processing on the facial skin model.The processed model is then imported into Workbench.For different displacements,the driving point setting,constraint position are studied.By analyzing the simulation results,the optimal displacement of the driving skin was finally obtained,and six facial expressions were simulated.The mold is made of stucco,and a food-grade silicone material is used as a raw material to complete the facial skin.According to the driving principle of steering gear,SMA and IPMC,build a control platform to complete the independent motion control of the neck,eyes,jaw and eyebrows;test the prototype of the robot and use multiple SMA drives to drive the facial skin.On this basis,the robot's six facial expressions were experimentally studied.Finally,the conversion experiments between different expressions and the voice control expression experiments were completed.