Adaptive trajectory tracking of wheeled mobile robot with uncertain parameters

A wheeled mobile robot (WMR) belongs to the class of non-holonomic systems with highly nonlinear dynamics. Because of their fast maneuvering and energy saving characteristics, these robots are especially popular in following or tracking a pre-defined trajectory. The trajectory of a WMR is controlled with the help of two very different control schemes namely model dependent approach and model free approach. While the model dependent approach relies on a particular model for the controller design, the model free method controls the trajectory with the help of learning methods. A Direct Model Reference Adaptive Controller (D-MRAC) is described for the model based technique, while an Adaptive Neuro-Fuzzy Inference System (ANFIS) is used for the model-free adaptive control design. With the help of simulations, it is shown that data driven intelligent approach is comparable to model dependent approach in terms of tracking performance and therefore can be preferred over complex model dependent adaptive algorithms.