Research On Anthropomorphic Variable Stiffness Control Technology Of Virtual Hand And Robotic Arm Based On Surface Electromyography

Impedance control can determine the dynamic relationship between the interaction force between robot and environment and trajectory,which makes robots perform tasks or interact with environment more compliant.It is widely used in robot control.However,parameters such as stiffness should be determined according to task requirements,the effect of interacting with unstructured environment or performing some uncertain dynamic tasks is not deal.The learning-based variable stiffness control method can adjust the stiffness of robots,but it is not direct enough and will increase the cost of the robot system.Human beings can complete some complex operations skillfully and safely by adjusting the stiffness of human body adaptively.And surface Electromyography(sEMG)contains much information about the muscle strength,joint motion and joint stiffness.Referring to the excellent operation ability of the human body,the stiffness adjustment features of human body are extracted from sEMG,and the anthropomorphic variable stiffness control technology of virtual hand and robotic arm based on sEMG is studied in this paper,and the following four aspects are mainly carried out:(1)A virtual hand interaction system based on sEMG and variable stiffness impedance control is designed.A method for estimating the joint angle of virtual hand from sEMG based on the variable stiffness impedance control model is proposed,and the stiffness of the virtual hand is adjusted dynamically according to the stiffness level of the human hand,which enables the virtual hand with compliance during interaction.The virtual hand interactive force feedback method based on the wearable fingertip force feedback device is studied,which enhances the operator's immersion during interaction by feeding back the interactive force of virtual hand to the operator.(2)The experimental research of virtual hand based on sEMG and variable stiffness impedance control is carried out.First,the virtual hand angle estimation experiment was carried out,which verifies the effectiveness of the virtual hand angle estimation method based on sEMG and variable stiffness impedance control.Then the virtual hand interactive experiments were carried out in two virtual interaction scenarios,which are the grasp experiment and the anti-disturbance trajectory tracking experiment.The experimental results show that the virtual hand based sEMG and variable stiffness impedance control can adjust the stiffness by muscle contraction according to the task,which generates smaller interaction force on the premise of successfully grasp in the grasp experiment and resists disturbance effectively when tracking the trajectory in the anti-disturbance trajectory tracking experiment,so as to improve the flexibility of virtual and and the adaptability to different types of tasks.(3)A master-slave robotic arm system based on sEMG hand anthropomorphic variable stiffness control is designed.The mapping relationship between Omega.7 haptics device and UR3 robot arm is established by coordinate transformation,which realize the master-slave control.A robotic arm anthropomorphic variable stiffness control strategy based on sEMG is proposed.The stiffness level of human arm is extracted from sEMG of the operator in masterend to adjust the stiffness of robotic arm in slave-end.And the reference trajectory of corrected by impedance control according to the force,which makes the robotic arm have both position accuracy and compliance when interaction with objects.(4)The robot arm experiment research based on sEMG and anthropomorphic variable stiffness control was carried out.The UR3 robot arm experimental platform was built.And the UR3 robotic arm interactive experiment with fixed stiffness and UR3 robotic arm interactive experiment based on sEMG and anthropomorphic variable stiffness control were carried out respectively.The experimental results show that the low stiffness robotic arm has low position control accuracy and high compliance while the high stiffness robotic arm has high position control accuracy and low compliance.The UR3 robotic arm based on sEMG and anthropomorphic variable stiffness control has the higher position control accuracy compared with low stiffness and higher compliance compared with high stiffness,which can adapt to unknown environment and tasks with uncertain.