Interval type-2 fuzzy neural network-based adaptive compensation control for omni-directional mobile robot

In order to overcome the influence of model uncertainty and external disturbance on the trajectory tracking accuracy of four-wheel omnidirectional mobile robot (FM-OMR), a new adaptive trajectory tracking control scheme based on interval type 2 fuzzy neural network approximator (IT2FNNA) is proposed in this paper. Based on the kinematics and dynamics model of FM-OMR, a dual loop trajectory controller is constructed. To improve the adaptive ability of IT2FNNA, an adaptive adjustment method for the network output is proposed. In addition, to improve the approximation accuracy of the IT2FNNA, a bias strategy is proposed based on the FM-OMR model. Finally, the circle track and "8" track are simulated and tested. The simulation and experimental results show that the bias strategy and adaptive factor really improve the approximation ability and accuracy of IT2FNNA, and the proposed adaptive control scheme has smaller fluctuations, higher accuracy and better steady-state performance.