Inverse ACO Applied for Exploration and Surveillance in Unknown Environments

摘要

This paper focuses on a distributed strategy proposed to coordinate a multiple robot system applied to exploration and surveillance tasks. The strategy is based on the artificial ant system theory. According to it robots are guided to unexplored or not recently explored regions. The main features of the strategy are, among others: low computation cost; and independence of the number of robots. Results from preceding investigations confirm the strategy is able to emerge a cooperative robot behavior, that is, the exploration and surveillance tasks are synergistically executed. This paper concerns specifically the robustness of the coordination strategy regarding to the environment structure. Two metrics are adopted for evaluation: needed time to conclude the exploration task, and time between two consecutive senses on a same region. Simulation results show that the coordination strategy is able to establish effective trajectories, that is, robots are guided to explore the environment and to sense repeatedly and completely the environment.