Offensive Strategy in the 2D Soccer Simulation League Using Multi-Group Ant Colony Optimization

Abstract

The 2D soccer simulation league is one of the best test beds for the research of artificial intelligence (AI). It has achieved great successes in the domain of multi-agent cooperation and machine learning. However, the problem of integral offensive strategy has not been solved because of the dynamic and unpredictable nature of the environment. In this paper, we present a novel offensive strategy based on multi-group ant colony optimization (MACO-OS). The strategy uses the pheromone evaporation mechanism to count the preference value of each attack action in different environments, and saves the values of success rate and preference in an attack information tree in the background. The decision module of the attacker then selects the best attack action according to the preference value. The MACO-OS approach has been successfully implemented in our 2D soccer simulation team in RoboCup competitions. The experimental results have indicated that the agents developed with this strategy, along with related techniques, delivered outstanding performances.

An Improved Ant Colony Optimization Algorithm Based on Hybrid Strategies for Scheduling Problem

“…Krynicki et al [50] proposed a multi-pheromone variant of ACO system. Chen and Wang [51] proposed a novel offensive strategy based on multi-group ACO for the 2D soccer simulation league. Marzband et al [52] proposed a multi-layer ACO algorithm for real time experimental implementation of optimum energy management system.…”

Стан та перспективи розвитку командної взаємодії роботів на прикладі змагань світового турніру «RoboСup»

“…Visual information is sent an average of 6-7 times per second. During a standard ten-minute game, the agent receives visual information approximately 4,000 times and is able to respond 6,000 times [18].…”

“…The first three actions (kicking the ball, passing, dribbling) are attacking, the other four (blocking, capturing, marking, formation) are defensive, while the middle two actions (positioning, interception) can be used as an attack and in defense for players who do not own the ball. The actions are defined as follows [18]:…”

A Dynamic Mining Algorithm for Multi-granularity User’s Learning Preference Based on Ant Colony Optimization

“…Then ants tend to move in the direction of high pheromone concentration. Thus, the group behavior of the ant colony shows a positive feedback: the more ants travel on a path, the more probability the latter choose the path, so, the path with good quality and short distance will attract more ants, and the pheromone concentration grows faster [14].…”