Trajectory Tracking Control And Simulation Of Multi-degree-of-freedom Manipulator

Multi-degree-of-freedom (MDOF) manipulator is an important component of robot. It applies widely in human production. In this paper the solution of inverse kinematics and trajectory tracking are emphatically analyzed based on manipulator kinematics and dynamics. And take six degrees of freedom manipulator as an example, its kinematics, dynamics and trajectory planning are simulated.As the solving of manipulator inverse kinematics, we can use the genetic algorithm or back propagated algorithm to train the weights of neural network then obtain the solution of the problem. However, these methods need to be further improved in solving accuracy and convergence rate. Through the similarity between radial basis function neural network and immune principle, use the generalization ability of artificial immune principle to RBF network data sets training, and adjust the structure of the neural network hidden layer. And then use recursive least squares method to determine the weight value of network. So as to adjust and learn the structure and weight value of the neural network self-adaptively. The simulation results of manipulator verify that the trained neural network by the immune principle has better generalization ability and fast convergence. It can improve the solving accuracy of manipulator inverse kinematics.As an important aspect of robot control science, in the process of control system operation trajectory tracking has some uncertain factors such as system errors, signal detection errors, high-frequency characteristics, joint friction and so on. Traditional control method based on object is difficult to control the trajectory tracking precisely and unable to obtain satisfactory control performance. In this paper, combine fuzzy control and neural network for the simulation of RBF neural network and fuzzy system and use RBF fuzzy neural network algorithm to control the manipulator. In this algorithm, use fuzzy control to regulate the parameters and the structure of RBF neural network. And Nearest Neighbor Clustering Algorithms was adopted to renewal the fuzzy rules and optimize the neural network hidden layer neuron number so as to adjust the controller parameters and structure adaptively and accelerate the network training speed. The simulation results show that compared with traditional neural network algorithm, the improved algorithm has better performance, good learning speed, tracking accuracy and self-learning ability. Then in this paper, the forward kinematics, inverse kinematics, forward dynamics, inverse kinematics and trajectory planning of the six degrees of freedom manipulator Puma 560 are analyzed and get its corresponding simulation results.