Cooperative genetic algorithms: a new approach to solve the path planning problem for cooperative robotic manipulators sharing the same work space

Abstract

A new approach for planning paths without collisions for two robots sharing the same work space based on a new kind of genetic algorithm that we have called cooperative genetic algorithms (CGA) is presented. Each robot is associated with a population of the CGA. The cooperation is carried out through an interaction between populations until they find collision-free paths for both robots. The fitness function of the CGA considers reaching the goal without collisions by using a minimal number of robotic configurations. The main advantages of this approach are: it is not necessary either to use the composite C-Space or a kinematic model based on a Jacobian matrix; this method can be used for both redundant and nonredundant robots.