Autonomous path following in mobile robots with nonholonomic constraints can be divided into two problems: first, selecting the tracking point, and then designing an appropriate controller to follow the selected point. When selecting tracking point, considering the kinematics of the robot as well as characteristics of the desired path is of considerable importance. For these purposes, a curvature-based point selection algorithm is first proposed for the car-like mobile robot with independent steering mechanisms. Each instant, the proposed algorithm finds a point which enables the robot to be tangent to the path at that specific point. Afterwards, in order to take into consideration characteristics of the path, a fuzzy adaptive curvature-based point selection algorithm is proposed. In this algorithm, in addition to the kinematic constraints, path characteristics are also considered in selecting the tracking point. This gives the robot the ability to show better performance when the path slope changes suddenly, resulting in less overshoot/undershoot around the desired path. The fuzzy adaptive curvature-based point selection algorithm is combined with a controller based on the Takagi–Sugeno fuzzy logic, such that the fuzzy adaptive curvature-based point selection algorithm selects the tracking point, while the Takagi–Sugeno fuzzy logic controller makes the robot follow the selected point. Finally, the fuzzy adaptive curvature-based point selection–Takagi–Sugeno fuzzy logic tracker is implemented on the robot, and the results are compared with a similar path-following algorithm. Obtained results show that for tracking a piecewise linear path, the steering activity and the following root mean square error decrease from 170.74° and 0.37 m for the conventional fuzzy controller to 63.37° and 0.09 m for the fuzzy adaptive curvature-based point selection–Takagi–Sugeno fuzzy logic controller, respectively.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.