| Because robot's mathematical model has serious nonlinearity and strong coupling character and many uncertain factors such as friction and laden diversification, it can not get satisfactory control results by using the traditional control method based on object model. Fuzzy control does not depend on the object mathematical model, and has great robustness, which using in the robot trajectory control can effectively overcome the bad influences of control coupling, nonlinear and parameter changes, and achieve better control results. However the ordinary fuzzy controller's parameters and control rules can not be changed once they have been established, therefore it can not be well adapted to the dynamic characteristics of the system changes or the impact of random disturbance.Genetic algorithm is a parallel and efficient global optimized search algorithm based on the natural evolution theory, it can transform the optimization problems into a process of biological evolution by simulating the evolution of biology, and obtain the optimal solution by taking the strategy which the fittest one survive. For genetic algorithms and fuzzy control have strong complementary characters to each other, in recent years, the fuzzy control combined with genetic algorithms has become a hot research.Through deeply studing genetic algorithm and fuzzy control theory, for overcoming the shortcomings and inadequacies of the basic genetic algorithm, an improved genetic algorithm called dynamic parameter adjustment genetic algorithm is presented in this paper. The paper discussed how to use it to optimize the conventional fuzzy controller's membership functions and fuzzy control rules, and to achieve more excellent control results. Then a fuzzy control optimized by this designed dynamic parameter adjustment genetic algorithm is applied to the Robot manipulator trajectory control. First of all,we establish a practical 5 DOF joints Robot manipulator entity model by the SimMechanics in MATLAB,then we use it to construct the Robot manipulator's fuzzy control system which is optimized by the dynamic parameter adjustment genetic algorithm. We conducted an simulation experiment to test the tracking effect of the robot manipulator's trajectory through the Matlab/Simulink, the experiment results show that the designed fuzzy control optimized by the dynamic parameters adjustment genetic algorithm can reduce the impact of complex system's model error, external interference uncertainties to attain precise trajectory tracking effect. The simulation results also show that this control algorithm has good dynamic and static characters, as well as great anti-jamming and robustness characters. |
PDF Full Text Request