A New Distance Diffusion Algorithm for a Path-Planning Model based on Cellular Automata

Cellular automata (CA) are bio-inspired approach that have been recently investigated to several applications including robotics. An improved model based on CA rules is proposed and evaluated for path-planning in autonomous robots. The objective is to build a short collision-free path from the robot's starting position to the target, trying to avoid unnecessary turns as much as possible. CA rules are used to enlarge obstacles, avoid their concavities and spread the distance from each cell to the target. The robot route is planned using the information of the distance of each free cell to the goal. Experiments were carried out to confirm the efficiency of the new techniques employed. Simulations with the navigation of a e-puck robot using the Webots platform have shown promising results confirming that the model is able to plan smooth and short routes.