Evolving Effective Multi-Robot Coordination Strategies for Dynamic Environments Using Cultural Algorithms

摘要

Simulated robotic soccer is frequently used as a test method for contemporary artificial intelligence research. It provides a real-time environment with complex dynamics and sensor information that is both noisy and limited. Team coordination between the robots is essential for success. Cultural Algorithm (CA) is a branch of evolutionary algorithms, and it is used in this research to teach software robots to play soccer by finding the best action to execute depending on its position on the field, and its relation to the nearest opponent. The action of each agent is encoded by an integer string that represents the action rules. Our agents played against a team of defenders from well-known teams in order to enhance their offensive capabilities. Agents developed good offensive abilities through team coordination processes supported by Cultural Algorithms. The simulation results are obtained using the well-known Robo-Cup soccer simulator. The results of this research suggest the effectiveness of the proposed method as well as indicating future research directions.