Research On Trajectory Planning Of Robotic Arm Based On Image Recognition

In recent years,the rapid development of mechanical arm has also promoted the construction of automatic wharf.The lock pin is always a bottleneck on the development road of wharf automation because it is difficult to identify and disassemble.This paper studies the theory and method of image recognition and trajectory planning of the robot arm,and completes the identification and disassembly of the lock pin respectively.The main research contents include:In view of the kinematics equation derivation of the mechanical arm,D-H(Denavit-Hartenberg)modeling method is adopted to establish the forward kinematics model,and the inverse kinematics model is derived by algebraic method,and the 3d simulation model of the mechanical arm is established by Matlab software to verify the accuracy of the forward and inverse kinematics equations.According to lock pin identification problem,apply CNN of deep learning to lock pin identification,use experimental data to verify the validity of the CNN model,in order to prevent the disappearance of gradient because of the deeper model.On the basis of CNN,residual connection is added,Res Net model is proposed and LRN mechanism is introduced to further improve the efficiency of feature extraction.Tests are carried out on the same data set to verify that the ResNet model have higher lock pin image recognition rate and lower misidentification rate than the traditional CNN model.The joint space method is selected as the research object in the robot arm trajectory planning problem,and then the cubic polynomial and the fifth-order polynomial method are used to simulate in MATLAB software respectively.For the problem of unsmooth trajectory,the two-breaker midpoint cubic polynomial trajectory planning method is proposed first.And verify its feasibility.Finally,the fifth-order polynomial with multisegment midpoint is proposed to improve the smoothness of the trajectory and reduce the impact.After the constraints are given in the simulation experiment,the trajectory curves of the six joints are relatively smoother.