Robot positioning of a flexible hydraulic manipulator utilizing genetic algorithm and neural networks

Robot positioning requires that the actuator positions are calculated as a function of end effector position. This mapping is called inverse kinematics of a robot. The inverse kinematics problem is very nonlinear and in some cases it cannot be solved in closed form. Several iterative and neural network approaches are studied in solving the inverse kinematics problem. Deflection of the manipulator arms due to flexibility and mass load causes positioning error. The magnitude of the error depends on the amount of mass load and arm positions and the stiffness characteristics of arms. In this paper a method based on genetic algorithm is used to solve the inverse kinematics of a three degrees of freedom log crane. Neural networks are used to solve the correction values for deflection compensation