Advanced Connectionist Control Algorithm for Robotic Compliance Tasks based on Wavelet Network Classifier

In this paper, a new comprehensive intelligent control strategy based on connectionist learning of robotic system uncertainties and wavelet network classification of unknown robot environments is proposed. The proposed wavelet neural network classifies characteristics of environments, determines the control parameters for compliance control and in coordination with basic learning compliance control algorithm reduces the influence of robot dynamic model uncertainties. In order to verify the proposed approach, compliant motion simulation experiments with a robotic arm placed in contact with a dynamic environment are realized. Computer simulation shows that the neural network classification provides a significant force error reduction.