Research On Trajectory Planning And Compliant Control Of Industrial Robot Based On LinuxCNC

Industrial robots refer to multi-joint robotic arms used in the industrial field,and the development level of industrial robots is an important standard to measure the level of manufacturing in a country.As a manufacturing country,China's research on robotics are of great significance for accelerating the transformation and upgrading of the manufacturing industry,improving the international competitiveness of the manufacturing industry,and accelerating its entry into the ranks of manufacturing powers.Trajectory planning is an important part in robotics,which is related to the performance and accuracy of industrial robots.Therefore,research on trajectory planning is indispensable for improving the performance of industrial robots.As industrial robots inevitably encounter collisions during their work,considering the safety of industrial robots and workers,robots should have the ability to adapt to external forces.The most common used method to make robots develop compliant motion when subjected to an external force is impedance control,which is also studied in this paper.The main contents of this paper are as follows:Firstly,the open-source software Linux CNC is studied,and the task controller and motion controller in Linux CNC are analyzed in detail,which provides a theoretical basis for the realization of algorithm that proposed later in this paper.Secondly,the B-spline trajectory planning in Cartesian space is studied,and the instantaneous velocity interpolation method is introduced to interpolate the B-spline.The experiment shows that,Compare to the equivalent distance interpolation method,using this method to control the position error in the interpolation process is more convenient.This paper also proposed a planning method which combining B-spline curve and improved particle swarm optimization algorithm in joint space to solve time-optimal trajectory planning problem.The improved particle swarm algorithm uses variable learning factors instead of constant learning factors to avoid the problem of easily falling into local solutions,and uses rollback technology to maintain the diversity of the particle swarm.Simulation results show that using this method can generate a time-optimal trajectory for industrial robot.thirdly,the impedance control of industrial robot is studied.For the robot without force sensor,a new kind of impedance control is proposed.The desire impedance of the proposed impedance control is composed of damper and spring.The newly proposed impedance control does not require there is a force sensor on industrial robots.Experiments that compare the traditional impedance control and the newly proposed impedance control are conducted in Matlab/Simulink.the results show that the proposed impedance control can make the robot develop a compliant motion when subjected to an external force,but the compliance achieved by this method cannot meet the desired impedance standard.Finally,on our robot experimental platform,related experiments were designed and carried out.Experimental result verified the effectiveness of the proposed method.