Neural network controller for 6-PRRS parallel robots

A neural network controller has been designed to control 6-PRRS parallel robots. This controller eliminates the nonlinearity and coupling of previous controllers by using a distributed control strategy and a compound orthogonal neural network (CONN). It learns the degree of uncertainty in positional data and uses it as feedforward to compensate for tracking errors. In this way the system converges as quickly as possible. The stability of the system is achieved by a proportional-integrate-derivative (PID) controller in the feedback path, which ensures rapid and steady trajectory tracking of the 6-PRRS parallel robot. The controller is designed in discrete-form, with a simple structure that can be easily applied to other engineering problems.