Characteristics of a genetic based approach to path planning for mobile robots

This paper presents a novel intelligent path planning algorithm for mobile robots operating in cluttered environments. The formulation of the problem for solution by the genetic algorithm is described and experiments are performed to determine the most appropriate path representation. A novel path guidance mechanism termed displacement to goal minimization is described. Further, the approach proposed utilizes multiple constraints, the development of which is presented. The results show that utilizing the formulation proposed, the genetic algorithm can plan and optimize paths without placing restrictions on the shapes of the obstacles in the working environment. It is also shown that the best path optimization takes place when the path representation has a relatively large and flexible step size as well as a flexible direction change capacity.