Direct Computation of Robot Inverse Kinematic Transformations Using Hopfield Neural Network

Computation of inverse kinematic transformations is an important problem in the field of robotics as it must be solved in real-time in order to position the end-effector at the desired position in a given task-space. However, it is a difficult problem for it involves the determination whether or not at least one mathematical set of robot joint angle values exists that will produce the desired coordinate configuration. The mathematical solutions should be checked against the physical constraints associated with the manipulator. The advent of artificial neural networks has made it possible to obtain general learning schemes which can be used to arrive at feasible solutions to inverse kinematics problem in a constrained environment independent of a robotic structure. The intent of this paper is to use the Hopfield neural network for the direct computation of inverse kinematic transformations of two- and three-linked robots, thereby avoiding the off-line training of neural networks.