Gaussian Kernel Controller for Path Tracking in Mobile Robots

This paper describes the design of a Gaussian kernel based path tracking controller for mobile robots. In order to achieve successful navigation under hybrid navigation architectures, it is critical for the robot to follow the path provided by a highlevel planner even while moving between waypoints. This is particularly difficult in real life situations involving robot motion in challenging terrains. Existing controllers for this purpose such as the pure pursuit does not ensure smooth motion of the robot or accurate tracking while moving between path segments. This paper describes the design of a controller that can ensure accurate path tracking even in the presence of disturbances, by utilizing the gradients of moving Gaussian kernels. In order to characterize the performance of the proposed controller, two different sets of simulations are conducted. Based on the results of the simulations, the Gaussian kernel controller ensures accurate tracking of the provided reference path while addressing the shortcomings of existing controllers. The paper concludes with a discussion on future directions for improvement.