A New High-Precision DZNN Model for Motion Control of Redundant Manipulators with Obstacle Avoidance

Aiming at the motion control and obstacle avoidance of redundant manipulators in practical applications, a continuous zeroing neural network (CZNN) model is first developed. For convenient digital computation, a new high-precision discrete zeroing neural network (DZNN) model is put forward on the basis of a three-step Adams-Bashforth formulation, which is termed fourth-order discrete zeroing neural network (FDZNN) model. Meanwhile, some conventional discrete models are also presented for comparison. The effectiveness and superiority of the FDZNN model are verified by conducting simulation experiments on the platform of planar six-axis manipulator.