Global path planning with obstacle avoidance for omnidirectional mobile robot using overhead camera

Abstract

Path planning is one of the indispensable modules of autonomous mobile robots that delineates a collision-free path between two desired positions in an obstacle-cluttered workspace. In this paper, we propose a global path planning method in the image plane using a single overhead camera based on the principle of artificial potential fields. Our algorithm optimally fuses an image-based technique for obstacle avoidance with path planning in the image space and integrates the CAD-based recognition method. The proposed method is suitable for planning the desired path for the omnidirectional mobile robots, and it is used as an input to our previously developed path-following controller. Experiment results show the efficiency of the generated path using an overhead camera for the omnidirectional robot iMoro which is a four-wheeled, independently steered mobile robot.