Research On Intelligent Trajectory Planning Of Parallel Manipulator Basted On Binocular Vision

Delta parallel manipulator has many advantages,such as simple structure,strong stiffness,strong load-carrying capacity and high positioning accuracy.In this paper,the trajectory planning of three-degree-of-freedom Delta parallel manipulator is studied.In the actual working environment,the manipulator needs to have the ability to perceive the mobile operation,so this paper introduces the binocular camera to collect the three-dimensional information of the manipulator workspace.Firstly,the overall structure of Delta parallel manipulator is introduced,and the forward and inverse kinematics models of the parallel manipulator are established.The accuracy of the kinematics model is verified based on Matlab software,and the trajectory of the manipulator is described in three-dimensional space.Secondly,the B-spline interpolation algorithm is used to plan the path of the parallel manipulator.the optimization index based on the weighting of the optimal time and the optimal trajectory smoothness is the fitness value of the PSO algorithm,and the global optimal particle is found in the population.The global optimal particles are searched in the population to complete the optimization of the manipulator based on spline interpolation path,and the simulation verification of the trajectory planning of the manipulator is completed under the condition of adding the actual constraints.Next,it mainly studies the key technologies involved in the process of 3D pose acquisition of workpiece based on binocular vision.It includes binocular camera calibration and correction based on OpenCV,binocular stereo matching,3D reconstruction and related image processing technology.Through the above steps,the 3D pose acquisition of workpiece based on binocular vision is realized,so that the manipulator is made to grasp according to the three-dimensional coordinates of the obtained workpiece.Finally,the hardware circuit design and the corresponding software design of the task allocator with TMS320F28335 as the core processor are carried out.At the same time,the upper computer software is developed,the human-computer interaction system is established,and the corresponding experimental debugging is carried out.The feasibility of intelligent trajectory planning based on binocular vision PSO optimization algorithm is verified by the experiment of the whole system of parallel manipulator.It is proved that the PSO algorithm based on time optimization and pulsating optimal weighting has a good effect on the actual running time and trajectory smoothness of parallel manipulator,which lays a foundation for further research and industrial application of manipulator trajectory planning.