Research On Modular Joint Flexibility Mechanism Based On Dual Feedback

In order to realize the intelligent manufacturing of 3C electronics,medical research,precision machining and other research areas production lines characterized by “flexibility,flexibility and change”,a cooperative robotic arm with force feedback capability emerged.Currently,the common force feedback method is installing variety of force/torque sensors at the end or joint of the robot,which not only increases the complexity of the robot structure,but also reduces the stability of the robot.Based on the modular concept,this paper designs a joint with high load-to-weight ratio,and uses the modular joint as the research object to establish a joint dynamics model and analyze the joint stiffness change.In this paper,the joint torque feedback method based on joint input and output dual position feedback is proposed,and the method is applied to the joint forcefree control research to realize the experiment of the load arm drag examination.The main research contents are as follows:First,the collaborative robotic arm joint design based on the modular concept.The positioning accuracy and load capacity are the two most important performance indicators of the collaborative robotic arm joint.This article accurately selects joint components based on the performance requirements of the robot arm.This paper proposes a hollow nested shaft joint design based on joint performance requirements,and analyzes the feasibility of the scheme.The article establishes a 3D model of the cooperative manipulator with modular joints,optimizes the joint structure,and develops a joint test platform.Second,modular joint dynamics modeling based on harmonic drive.The article uses the Newton-Eulerian modeling method to establish a dynamic equation that considers joint flexibility.This paper analyzes the influence factors of joint stiffness variation,studies the torsion,stiffness and error variation characteristics of harmonic transmission,and carries out transmission error simulation.The simulation results reveal the nonlinear variation of joint dynamics model based on harmonic drive.As a result,a modular joint dynamics model incorporating harmonic drive was established.Third,the joint parameter identification method based on the least squares method.Based on the joint dynamics model,the Stribeck friction model and the harmonic drive threestage stiffness variation model are used to independently identify the joint damping and joint stiffness.It is verified that the two models can approximate the joint friction and stiffness under low speed load conditions.Based on joint current and input-output dual-position feedback,the least-squares identification algorithm is used to identify the offline dynamic parameters of the joint dynamics model.In this paper,by analyzing the identification results,it is determined that the joint stiffness obtained by considering Coulomb and viscous friction is closer to the actual joint stiffness value.Finally,the joint forcefree control research based on torque feedback.According to the identification results of joint parameters,a method for estimating the stiffness of joint input and output with dual-position feedback and joint identification stiffness is proposed.The article sets up a comparative test,and the results show that the method can estimate joint torque with high accuracy under low speed load.The torque estimation method is applied to realize the joint torque feedback,and the joint friction force and the load arm weight torque are independently compensated,and the joint forcefree control in the torque mode is realized.Based on the accuracy of joint motion and the key point of joint compliance control,this paper studies the torque feedback method of forceless sensor.The study provides theoretical support for joint compliance control and is a useful addition to accelerate the next generation of collaborative robotic arm research.