Intelligent Agents for Observation and Containment of Malicious Targets Organizations

Abstract

The problem addressed in this work is an extension of the Cooperative Multi-Robot Observation of Multiple Moving Targets Problem (CMOMMT). The scenario remains the same, but the targets are structured as an organization to achieve the highest possible percentage of exploration of the environment and avoid robots. Targets can be organized as hierarchy, holarchy, team, and coalition, but they can also be unorganized. Our work seeks to apply computer vision to assist robots in classifying the target team’s organizational structure faced with a group of malicious target agents. Thus, robots can select the most appropriate strategy among the containment strategies implemented for each organizational structure or continue with the method proposed by literature for cases where the targets are not organized. The results showed that our approach had satisfactory results since, in luck, robots have a 20% chance of hitting the structure (hierarchy, holarchy, team, coalition, or random). Our approach had an accuracy of 63.28%. The containment strategies obtained satisfactory results in the robots’ performance regarding the depreciation of the Percentage of the Environment Explored by the Targets (PEET) compared to the previous approach for robots. However, for the Average Number of Observed Targets (ANOT), the previous strategy was better. The new organizational approach to targets in CMOMMT was better than random in the desired exploration of the desired environment.