A Spring Self-locking Nut Assembly System Base On Active Compliance Control

With the continuous development of modern industry,the application scenarios of industrial robots are becoming more and more abundant,which means that robotics is facing more and more requirements and challenges.Assembly robots are industrial robots used in assembly production lines to assemble parts.Compared with painting robots,mobile robots,etc.,assembly robots often need to control the contact force with the environment to complete assembly tasks.At present,the most widely used position control robots usually cannot control the contact force with the environment.Therefore,the research and application of compliance control is one of the urgent problems to be solved in the development of assembly robot technology.In this paper,a special assembly robot is designed for the assembly task of a new type of spring self-locking nut,and in-depth research is carried out around its mechanical structure design and compliance control algorithm development.(1)In this paper,the goal of spring self-locking nut assembly task is analyzed in detail,and the assembly task is refined.The functional requirements of the assembly system are described in detail from four aspects:the robot's degree of freedom,workspace,clamping task,and control algorithm.It points out the direction for the subsequent mechanical structure design and control algorithm design of the assembly system.(2)An assembly robot with a kinematic configuration of PRP type is designed according to the degree of freedom requirement of assembly system.In response to the requirements of work space and clamping task of the assembly system,key components such as joint drivers and sensors are selected,and non-standard parts such as end-effecter and connectors are designed.Aiming at the problem of sudden force change in the contact between rigid bodies,the structure of the end-effecter is optimized,the contact object during assembly is changed from "end-effecter-nut" to"spring-nut",and the environmental stiffness is reduced from 105N/mm to 7N/mm,thus realizing the passive compliance of the designed robot.Then,the 3D design of the assembly robot and the fabrication of the physical object are completed,and the kinematic model is established by using the Modified D-H method.Finally,through the calculation and analysis of the assembly robot's load capacity,kinematics model and workspace,the feasibility of its structural design in the assembly task of spring self-locking nut is verified.(3)According to the control algorithm requirements of the assembly system,the force/position hybrid control algorithm is used for constraint analysis,and the control mode of each joint stepping motor is determined.Then control algorithms are designed for each stepping motor.For the stepping motors of the first joint and the second joint,the position control algorithm is designed based on the increment PID control using the feedback value of the corresponding sensor.For the stepping motor of the third joint,the pressure sensor is used to feed back the contact force between the end and the environment,and then the position-based impedance control model is discretized,the force control algorithm is designed to realize the active compliance of the designed robot.Finally,a simulation model of position-based impedance control is established by Simulink,and the influence of impedance parameters on the control system is summarized through the simulation results.Compared with the simulation results of direct force feedback control and PI control,the results show that the position-based impedance control has better control effect in the assembly system model in this paper.(4)Aiming at the problem of time-consuming and inefficient when manually tuning impedance parameters,an impedance parameter optimization method based on hybrid particle swarm optimization is proposed.First,the 3D model of the assembly robot is simplified and imported into Adams.And through the co-simulation with Simulink,the contact model between the end-effecter and the nut of the assembly robot is established.Then,the fitness function with force overshoot and adjustment time as the optimization objective is designed,and the damping coefficient Bd and stiffness coefficient Kd of the position-based impedance controller are tuned using the hybrid particle swarm optimization.The simulation results show that the optimized position-based impedance controller can not only make the end-effecter to produce smaller force overshoot and steady-state error when in contact with the environment,but also quickly reach a steady state.And compared with the manually tuned impedance parameters,the optimized impedance parameters can achieve better control effect.