Interval Type-2 Fuzzy System for Autonomous Navigational Control of Non-holonomic Vehicles

This paper reports the strategy implementation, in simulation environment, for navigation of non-holonomic vehicles using an Interval Type-2 (IT2) Fuzzy Logic System (FLS). A Graphical User Interface (GUI) in MATLAB has been designed to aid the simulations of navigation starting from any position and orientation in a designated area to a designated parking place and fixed orientation. The FLS takes two inputs, vehicle location and orientation w.r.t. x-axis, to deliver one output, the steering angle, relevant to the steering process. The variables take on linguistic values for invoking the rulebase to deliver decisions to steer the vehicle to reach its final parking position. Linear Path Approximation (LPA) trajectory-traversing algorithms for non-holonomic motions have been integrated into the design of the GUI. The FLS was originally developed using Type-1 Fuzzy Sets (T1 FSs), and then enhanced to IT2 FLS by making use of IT2 FSs to handle uncertainties of T1 FLSs. The response of IT2 FLS becomes equivalent to that of T1 FLS as the extent of uncertainty, i.e., Footprint of Uncertainty (FOU) is reduced to zero. However, for restricted FOU within certain bounds, a considerable improvements is observed over the case with zero FOU, i.e., compared to T1 FSs case.