Three Joint Robotic Arm Trajectory Controllable

Mining robot will be an indispensable tool replacing person operatingunder heavy, dangerous and harsh environment, and also is the key technologyand equipment of the focus of the state development. System of mining robot isnot only a very complex time-varying, strongly coupled, highly nonlinearsystem, and in fact there are still many uncertainties, so we can not gaincomplete and accurate model of robot system. From the technical point ofcontrol excavator in order to make mining robot smarter and moreenvironmentally friendly, we will study the motion control of manipulator.On the basis of understanding of the mining robot device this paper takingthe three joint mechanical arm as research object systematically study thetrajectory control of three joint mechanical arm. In the course of the study, theestablishment of System model used some mathematical knowledge includingkinematics and dynamics, the trajectory control applied control theory includingneural networks and fuzzy control.With the help of the mathematical knowledge and modeling methods ofrobot kinematics and dynamics, this paper established kinematics model of threejoint mechanical arm and dynamics model based on Lagrangian function, at thesame time analyzed related issues and laid a theoretical foundation for thefollowing research trajectory planning and the path tracking control. Then, thispaper had a brief introduction to the trajectory planning and related issues and amore detailed analysis of the building process of the five sections of cubicpolynomial interpolation function in the manipulator joint space,derive theequation expression.On this basis, this paper planned the trajectory of Cartesianjoint path for the mechanical arm.Solving inverse kinematics of mechanical arm is an important step ofcarrying motion control of mechanical arm, this paper used that neural networkshad the ability to approximate arbitrary nonlinear system and studied theapplication of the neural networks of small brain in solving inverse kinematicsproblem of mechanical arm. The inverse kinematics problem is very complex,and there are the things that the inverse solution may not exist and the existing inverse solution may not be unique, here we studied the inverse kinematicssolution of three joint mechanical arm with the use of the CMAC network. Thesimulation result shows that the neural network can effectively solve the inversekinematics problem of mechanical arm, the solution reached a high precision,and also the solving time was short, so the neural network met the requirementsof real-time control of mechanical arm.On the basis of reference to large number of domestic and foreign literatureand against the trajectory tracking control problem of three joint mechanical arm,the introduction of intelligent control methods in this paper made the systemachieve good control. Taking into account the complexity of the path trackingcontrol, in order to achieve accurate trajectory tracking motion control, thispaper presents a RBF fuzzy neural network control algorithm which combinedneural networks with fuzzy control, neural networks and fuzzy control had theirown advantages and they were complementary. After combination RBF fuzzyneural network had fuzzy logic reasoning ability which ensured the stability ofthe system and improved the dynamic performance of the system andexperimental results showed good tracking performance and control effect.Study finding we obtained from motion control of the three jointmechanical arm has an important reference and significance to other structuralforms of robot, this will be a strong impetus to the forward development of robottechnology, expand the field of robotics research, more and more widely solvethe problem which can not be resolved by manual operation and provide moreconvenience for human life.