The Robot Teleoperation Control Based On SEMG

With the development of technology,the robotic teleoperation technology plays an important role in the fields of industry,aerospace,and ocean exploration,medical operations and services.The traditional teleoperation technology can just control the position of the remote robot,such as using keyboard and handle to teleoperate the robot arm.Although the operation method is simple,the interactivity between the human operator and the remote robot is poor.Take in account the arm information of human operator,including joint angles and the stiffness information,which can be used to improves the safety and trajectory tracking accuracy of the manipulator in different working environments.The robotic arm with flexible joint are widely used in industry and service industries due to its unique properties.However,because of the nonlinear of the control system and the uncertainty of inertial parameters,the robot arm with flexible joint is difficult to build model in practical applications.Moreover,the robot often works in unknown environment and there are various external disturbances applying in the robot during the teleoperation.How to model of the robot arm to ensure the tracking accuracy and effectively reduce external disturbance in remote operation,which is the most important issue to be solved in this paper.The surface Electromyography(sEMG)signals can be used to estimate the stiffness of human arm in some extent,the collected sEMG signals is non-invasive and simple opeerate.The human joint angles and stiffness of arm estimated by the sEMG signals are used as control inputs,and the Baxter robot with flexible joints is applied in the paper.The teleoperated control system and the method of reducing disturbance is studied,which mainly include:1)By analyzing the shortcomings of the method that mobile robot in obstacle avoidance under the classical potential field method,which is easy to fall into a local minimum.A teleoperation obstacle avoidance method based on sEMG is proposed.It is a solution to the problem that mobile robots can easily fall into a local minimum when the target point,obstacles and robot movement direction are on the same straight line.2)The human arm is modeled as a robot arm and the joint angles of the operator's arm are calculated.According to the standard DH parameter method,the operator's arm included shoulder joint and elbow joint,is regard as a 5-degree-of-freedom(DOF)chained robot arm.To calculate the joint angles,two wearable device are worn on the upper and lower arms of the operator respectively.3)Based on the known model of the robot arm,the disturbance observer is an effective technology for compensating low-frequency disturbance.However,the remote robot arm often works in the unknown environment,and the disturbance observer does not have the function of compensating interference when the high-frequency interference is applied on the robotic arm.By modeling and analyzing the sEMG signals of human arm,a variable gain controller based on the sEMG signals is designed to reduce the high frequency disturbance implied on the robot.4)For unknown parameters in the robot arm model,the radial basis function neural network(RBFNN)is used for approximation and the unknown model is converted into a known model.Finally,the experimental simulations represent the effectiveness of teleoperation control method based on sEMG signals in resisting interference.