Adaptive teleoperation control method based on RBF-Neural Networks and performance analysis

This work presents an adaptive teleoperation control method based on RBF-Neural Networks. First, the model of the teleoperation system with two slave robots is built. Then the controllers of the slaves and masters are designed separately. For the slave side, the dynamic uncertainties are considered as the main factor to influence the system stability, which is estimated by the RBF-Neural Networks (RBF-NNs). The structure parameters of the masters are known before the operation. Furthermore, we discuss the system stable conditions and position tracking effect of the slaves to the maters' motions. The proofs reveal that the system will converge to the stable states based on the assumptions that the estimating errors are smaller than a threshold value. The final tracking errors are corresponding with bounding values of the estimating errors of RBF-NNs method. Finally, a simulation is taken to certify the effectiveness of proposed method and the main conclusions.