Optimal task assignment for serial-parallel hybrid robots cooperationvia ant colony optimization

Abstract

An optimal task assignment method for a two robots cooperation system which consists of a serial robot Puma and a parallel robot Stewart is described in this paper. Both robots are with initially identical functionalities. A hierarchical control architecture is established for the system whose assigned task here is NP-hard. In the higher hierarchy, ant colony optimization (ACO) algorithm inspired by the behavior of natural ants is adopted to take charge task assignment for each robot, resulting in reliable and efficient division of labor; on the other hand, simple position control with kinematics is utilized for the lower hierarchy to perform task execution in dealing with the computation of expected joint angles according to appointed path points. Optimization is implemented by setting the goal of finding a best assignment strategy which successfully accomplish the task with the least cost of time. The effectiveness of the proposed method is validated through a new-designed mechanism in industrial spray-paint.