Numerical assessment of path planning for an autonomous robot passing through multi-layer barrier systems using a Genetic Algorithm

Because of the necessity of product transportation using an autonomous robot in industries, the path-planning in finding an appropriate way without collision is essential. The main purpose of this study is to solve the problem of robotic path-planning utilizing a Genetic Algorithm (GA), a robust scheme used in searching for the global optimum by imitating a genetic evolutionary process. In this study, a two dimensional mobile robot used in four kinds of multi-layer barrier systems (a one-layer barrier, a two-layer, a three-layer and a four-layer barrier system) has been introduced. To access a shortest path when an autonomous robot passing through multi-layer barrier systems, an objective function of the path length is minimized using a GA optimizer in conjunction with a minimum square root method as well, as a penalty function. The genetic algorithm provides a solid alternative to conventional methods of path-planning. Moreover, the optimization parameters for the desired path can easily be changed without a total overhaul of the overall algorithm. Results reveal that the path optimization within a limited working area can be simplified by presetting the fixed steps in an x-axis. Five kinds of GA parameters (pop, bit, iter, pc and pm) play essential roles in the solution's accuracy during GA optimization. Obstacle with more layer-barriers will increase the difficulty of short-path searching. An appropriate path can be obtained by increasing the iter from 500 to 5000 or 10000 during GA optimization process. Consequently, an efficient path that avoids multiple obstacles within a working area can be easily found using the GA algorithm. © 2010 Asian Network for Scientific Information.