Navigation design with SVM path planning and fuzzy-based path tracking for wheeled agent

A navigation system based on support vector machine (SVM) path planning and fuzzy sliding-mode controlled (SMC) path follower is developed for wheeled agent in this paper. The developed system, comprising image acquisition, map formatting, path planning, label assignment and path tracking inference mechanism, aims to gracefully follow a planned smooth path. In the first portion, a Voronoi diagram is employed as a preprocessor to roughly fit a safe route between the initial and goal positions with discontinuous fragmental segments to avoid obstacles. A postprocessor of Gaussian kernel SVM is then applied to consecutively convert the fragmental path to a differentiable continuous path for a safe and smooth passage off line. In the second portion, a path tracking inference mechanism based on fuzzy SMC is proposed subsequently to track the planned path on line. In the study, a practical validation framework is implemented to assess the performance of the proposed system. Beside the two-stage path planner and the fuzzy SMC path tracking inference mechanism, the framework devises further an image processing to precondition the inputs of the system. With the real devised system, a series of experiments are carried out and analyzed to confirm the expected performance. The experiments show a robust capability of the system for both path planning and path tracking under various obstacle layouts.