ANN-based sliding mode control for non-holonomic mobile robots

The purpose of the paper is to propose a neural network-based sliding mode control law for solving the trajectory tracking problem of mobile robots. Artificial neural networks (ANN) help us choose a proper sliding surface, which is time-varying. The weights of the ANN are changed according to an adaptive algorithm to control the system state to hit a user-defined sliding surface and then slide along it. The input parameters to the ANN are chosen as delayed outputs of the sliding mode controller and delayed output of the plant. The sliding surface is adapted such that convergence towards the path to be followed is guaranteed. A non-holonomic mobile robot as a practical example for the application of this control system is considered