Research On Drive Control And Motion Planning Of High Precision Collaborative Robots

The collaborative robots have been widely used in modern industrial manufacturing filed.With the development of industry,the higher work efficiency,precision,stability and autonomy of collaborative robots are required.In order to improve the overall motion performance of collaborative robots,the joint driver,trajectory planning and path planning algorithms are researched in this thesis,shown as follows:(1)In order to improve the precision of joint driver,the position-sensorless drive control method for brushless DC motors(BLDC)is introduced.Because the back electromotive force(EMF)is very small at low speed,a modified algorithm including the improved zero crossing point(ZCP)detection,angle compensation and speed calculation approaches is proposed.First,based on the mathematical model of BLDC,the line-to-line back EMF can be gained.Then,to get the ZCP signal,a low pass filter(LPF)with alterable cut-off frequency is used to reduce the disturbance of the line-to-line back EMF.Finally,based on the symmetrical characteristics of three-phase back EMF,a novel compensation algorithm including an open-loop and a close-loop is proposed to get the precise commutation signal.Moreover,the proposed speed calculation algorithm can promote the response rate and stability of motor by updating speed in time when the speed drops.Both the simulation and experimental results demonstrate the high stability and reliability of the proposed methods.(2)In order to satisfy the high efficiency and high precision of collaborative robots,this paper proposes a novel trajectory planning method.First,in Cartesian space,an improved velocity look-ahead control algorithm and the acceleration/deceleration hybrid algorithm based on the cubic polynomial are combined to construct the end-effector trajectory of robots.Further,the joint parameters can be obtained through the inverse kinematics solver.Then,in joint space,to achieve the smooth and stable motion of robot,the joint trajectories are constructed through the improved velocity look-ahead control algorithm and quintic B-spline.Finally,the proposed trajectory planning method is tested on a 4-DOF serial collaborative robot.The experimental results indicate that the collaborative robot achieves the high efficiency and precision through the proposed method.(3)To improve the autonomy of collaborative robots in complex environments,an efficient path planning algorithm is proposed based on Rapidly-exploring Random Tree Star(RRT*).First,the dynamic incremental distance scheme(DID)is proposed to promote the flexibility of obstacle avoidance by adjusting the incremental distance of RRT* according to the obstacle density of environment.Then,in order to prevent the RRT* from trapping in local minima and cover unexplored areas quickly,the subregion sampling scheme based on Gaussian distribution(SSG)is presented.Based on the proposed schemes,the improved RRT* can rapidly find a valid path by reducing the number of iterations.Finally,the simulation experiments based on various complex 2D environments and a 4-DOF serial collaborative robot in 3D environment are adopted to demonstrate the effectiveness and practicability of the improved RRT*.The results show that the proposed algorithms significantly improve the path planning efficiency of collaborative robot.