Grid Ant Colony Optimization Applied to a Multi-robotic Garbage Collection System

摘要

Garbage recycling and collection problem is an interesting problem that researchers are applying swarm intelligence algorithms to solve. Some previous approaches used particle swarm optimization, immune systems and ant colony optimization algorithms and achieved good results. Ant colony optimization is a well-known swarm intelligence algorithm that is normally used to solve computational problems which can be reduced to finding good paths in graphs. A multi-robotic system can be applied to solve this problem but it will need a control algorithm to accomplish the task. Applying the regular ant colony optimization algorithm to control the multi-robotic system is not a trivial task due to the graph representation needed. This work proposes modifications in the ant colony optimization algorithm that uses grid representation and applies the modified algorithm to solve this problem. The results showed a decrease of one order of magnitude in the number of iterations needed to solve the problem compared to the previous version of the algorithm. Considering the results the proposed algorithm showed to be able to control a multi-robotic system for the chosen problem.