Genetic algorithm based path planning and dynamic obstacle avoidance of mobile robots

Abstract

A simple path planning scheme is proposed for navigation of mobile robots while avoiding obstacles. In generating the goal directed dynamic path, the path planning scheme uses a genetic search algorithm whose coding technique speeds up the execution of genetic search for fast path generation. The fitness value of the generated paths is evaluated in terms of the safety from the obstructing dynamic objects and the distance to the goal position by the genetic algorithm. The execution time of genetic search is further accelerated by projecting the two dimensional data to one dimensional ones to reduce the size of search space.