Neural networks‐based adaptive robust control of crawler‐type mobile manipulators using sliding mode approach

Abstract

PurposeThe purpose of this paper is to present the radial basis function (RBF) networks‐based adaptive robust control for an omni‐directional wheeled mobile manipulator in the presence of uncertainties and disturbances.Design/methodology/approachFirst, a dynamic model is obtained based on the practical omni‐directional wheeled mobile manipulator system. Second, the RBF neural network is used to identify the unstructured system dynamics directly due to its ability to approximate a nonlinear continuous function to arbitrary accuracy. Using the learning ability of neural networks, RBFNARC can co‐ordinately control the omni‐directional mobile platform and the mounted manipulator with different dynamics efficiently. The implementation of the control algorithm is dependent on the sliding mode control.FindingsBased on the Lyapunov stability theory, the stability of the whole control system, the boundedness of the neural networks weight estimation errors, and the uniformly ultimate boundedness of the tracking error are all strictly guaranteed.Originality/valueIn this paper, an adaptive robust control scheme using neural networks combined with sliding mode control is proposed for crawler‐type mobile manipulators in the presence of uncertainties and disturbances. RBF neural networks approximate the system dynamics directly and overcome the structured uncertainty by learning. Based on the Lyapunov stability theory, the stability of the whole control system, the boundedness of the neural networks weight estimation errors, and the uniformly ultimate boundedness of the tracking error are all strictly guaranteed.