Research On Human-robot Flexible Cooperation And Interaction Technology Based On EMG Signals

In recent years,the manufacturing industry is meeting more and more personalized needs of the market.At the same time,the demand for worker safety guarantee in the production environment is also increasing.Therefore,compared with the traditional industrial robots,the demand for cooperative robots with higher safety performance and more flexible operation is increasing in the current production and manufacturing.Collaborative robot solutions can quickly adapt to the tasks of line transformation and renewal and can be quickly and flexibly deployed into traditional production processes.However,in the physical level human-computer cooperation and interaction task scenarios,cooperative robots have the problem of insufficient flexibility.Aiming at this problem,this paper presents a flexible man-machine collaboration and interaction strategy based on the electromyographic signal in order to improve the collaboration of flexible robots,are introduced in the process of the man-machine collaboration to estimate the stiffness at the end of the operator's arm and intention recognition strategy,by tracking human arm structure and its level of muscle activation,online estimate at the end of the arm stiffness.In addition,through the acquisition and processing of EMG signal,the neural network model is used to detect human motion intention.Furthermore,in order to better integrate the changes of the operator's posture and muscle into the human-computer cooperation process,a humancomputer cooperation framework based on variable admittance control is designed.Inspired by the adaptability of human body in human-to-human cooperation,a smooth stiffness mapping between the end stiffness of human arm and the joint stiffness of robot arm was designed,so that the robot arm could inherit the flexibility of the arm.The strategy of end stiffness estimation is further extended to full impedance parameter estimation,and the estimated end impedance parameters of the human arm are used to map the robot arm,and the impedance model of the operator is matched with the admittance model of the robot arm,so as to obtain the motion trajectory suitable for the interaction force.Furthermore,in order to improve the trajectory tracking performance of the end-effector of the robot arm,a neural network controller based on function approximation is designed to compensate the uncertain load and unknown dynamics.In order to verify its effectiveness,the proposed human-machine flexible collaboration and interaction strategy is deployed on the Baxter collaborative robot platform,and the experiment is carried out in a real environment.The experimental results on the platform show good performance and verify the effectiveness of the strategy proposed in this paper.In addition to analyzing the experimental results,this paper also discusses the advantages and disadvantages of the proposed method and the development direction of the proposed strategy in the future.