Analysis and real-time implementation of a radial-basis-function neural-network compensator for high-performance robot manipulators

System performance of robot manipulators with nonadaptive controllers might degrade significantly in the presence of structured or unstructured uncert ainties. In order to improve the system performance, a novel radial-basis-f unction (RBF) neural-network (NN) compensator is proposed. With the RBF NN compensator introduced, the system errors and the NN weights with large dis persion in the initial NN weights are guaranteed to be bounded in the Lyapu nov sense. The NN weights of the RBF NN compensator are adaptively tuned. S everal software-based controllers, including the computed-torque control (C TC) and a few RBF NN schemes, are implemented in an industrial manipulator in real time. Experimental results are obtained to demonstrate the relative effectiveness of the proposed controllers in improving the tracking perfor mance of the robot manipulators associated with structured or unstructured uncertainties. (C) 1999 Elsevier Science Ltd. All rights reserved.