Multiobjective Problem to Find Paths through Swarm Robotics

摘要

In this paper, we present a scheme to solve the problem to explore, cover and define paths in dangerous environments. The proposal is organized in three phases: exploration, mapping and path planning. Swarm robotics is used in the first phase along with a stigmergy approach for the communication between the agents. The explored cell will be stored in order to create a grid map. In the path planning phase, a multiobjective problem is used to minimize distance and danger, through a graph obtained by an adapted RRG, and a genetic algorithm called modified NSGA-II.