Modelingand Motion Controlof Multi-joint Series Robot

China is the great power of manufacturing industry and it changes to the powerful nation of manufacturing industry.Industrial robots develop a very important role.In the study of robot arms,kinematics is the foundation,motion planning is the expected trajectory source,and trajectory tracking is the core.Based on these three aspects,the author conducted corresponding study.Kinematics of robot arms: first of all,Denso VP6242 G robot was regarded as an object.The kinematical equation was derived.According to the kinematical equation,a traditional inverse transformation method was used to solve inverse kinematics.Meanwhile,correctness of forward and inverse kinematics was simulated and verified.Then,by aiming at problems unsolved by a traditional method for robot arm' inverse kinematics with general structure,the author proposed a solution of high-precision real-time inverse kinematics excluding inverse trigonometric.The solution was based on the modified PSOGSA.Simulation verification was used to verify accuracy,rapidness and uniqueness of the solution.At last,in tool coordinate calibration,by aiming at the large deviation in calibration of robot arms,the author came up with a rapid calibration method on the basis of multipoint and multi-pose estimation method,added the wrong judgment mechanism and verified algorithm correction through experiments.Motion planning of robot arm: on the basis of summarizing motion planning in general forms,the author discussed route planning to use artificial potential field to avoid obstacles automatically and conducted the simple obstacle avoidance simulation for the method.Meanwhile,a 5-3-5 SPL interpolation joint trajectory planning method was studied.Under the precondition of safeguarding continuous derivative of speed and accelerated speed,it also can ensure that the accelerated speed is continuous and starts and stops from zero.It is better to restrain vibration of robot arms than the conventional method.Matlab simulation and Denso robot arm could be used to verify effectiveness of the method.Trajectory tracking control of robot arm: By simplifying the six-axis kinetic model of Puma robot arm,front three-axis model with the back three-axis locking was obtained.Then,on the basis of summarizing and analyzing expected trajectory generation method of common joints,the author obtained the general structure for trajectory tracking control of robot arm and studied trajectory tracking control methods of PD,PD based on compensation gravity and PD based on factored moment.Based on it,the author proposed the neural network compensation control method of factored moment for the time-varying PD parameters to greatly improve the comprehensive control effects.