Multi-robot scheduling using evolutionary algorithms

An evolutionary and hybrid approach to the problem of routing and scheduling a team of robotic agents to perform a resource distribution task in a static environment is presented. The essence of the algorithm is in the implementation of a central planner responsible for planning the routes and schedules for the whole team of agents. The innovative genetic approach breaks down the task of multiple route design into a single traveling salesperson problem and then uses different combinations of genetic operators to converge to nearly optimal solutions to the transformed representation. The key advantage of this approach is that globally optimal or near optimal solutions can be produced. The results obtained on some of the standard problems are quite encouraging and near optimal route distributions were found using both approaches.