[1]付鹏,罗杰.基于改进蚁群算法的Q学习算法研究[J].计算机技术与发展,2013,(02):123-126.
　FU Peng,LUO Jie.Q Learning Algorithm Research Based on Improved Ant Colony Algorithm[J].,2013,(02):123-126.
点击复制

基于改进蚁群算法的Q学习算法研究()
分享到：

《计算机技术与发展》[ISSN:1006-6977/CN:61-1281/TN]

卷:
期数:
2013年02期
页码:
123-126
栏目:
智能、算法、系统工程
出版日期:
1900-01-01

文章信息/Info

Title:
Q Learning Algorithm Research Based on Improved Ant Colony Algorithm
文章编号:
1673-629X（2013）02-0123-04
作者:
付鹏; 罗杰
南京邮电大学 自动化学院
Author(s):
FU Peng; LUO Jie
关键词:
多Agent系统Q学习; 改进蚁群算法; 围捕问题
Keywords:
Q-learning algorithm in the multi-agent system; improved ant colony algorithm; round up problem
文献标志码:
A
摘要:
文中以围捕问题作为研究平台,以提高多Agent系统中Q学习算法的学习效率作为研究目标,提出了一种基于改进蚁群算法的Q学习算法.该算法将信息素的概念引入到Q学习中,结合采用动态自适应调整信息素挥发因子的蚁群算法,使Agent在进行行为决策时不再只以Q值作为参考标准,而是考量Q值与信息素的综合效应,加强了Agent彼此间的信息共享,增强了交互性.并且对于复杂变化的周围环境,根据具体环境条件,设立分阶段的多奖惩标准,使算法对于环境和状态有更好的适应性.仿真实验证明了改进后的Q学习算法提高了学习系统的效率,高效地实现了多Agent系统的目标任务
Abstract:
On the basis of round up problem,propose a Q-learning algorithm based on the improved ant colony algorithm in order to im-prove the learning efficiency of Q-learning algorithm in the multi-Agent system. The algorithm introduces the concept of pheromone to Q-learning process,combined with ant colony algorithm of dynamic adaptive adjustment volatile factor,so that the Q value is no longer the only determinant for decision-making of the Agent,but is considered to be a part of the combined effect with pheromone,which can strengthen the sharing of information and enhance the interaction between Agents. In addition,the Q-learning algorithm can set phased in-centive standards according to complex surrounding environment and specific conditions,achieve better adaptability of itself to the envi-ronment and conditions. The simulation experience results show that the improved Q-learning algorithm increases the efficiency of the learning system and achieves objectives of the multi-Agent system effectively

相似文献/References:

[1]方昕.一种求解高校路网的改进蚁群算法策略与应用[J].计算机技术与发展,2012,(12):142.
　FANG Xin.Strategies and Application of Improved Ant Colony Algorithm to Solve University Path[J].,2012,(02):142.
[2]殷 星,魏 明.基于改进蚁群算法的 PTN 网络路径优化[J].计算机技术与发展,2020,30(12):83.[doi:10. 3969 / j. issn. 1673-629X. 2020. 12. 015]
　YIN Xing,WEI Ming.Path Optimization of PTN Network Based on Improved Ant Colony Algorithm[J].,2020,30(02):83.[doi:10. 3969 / j. issn. 1673-629X. 2020. 12. 015]
[3]顾文斌,杨生胜,王贤良,等.基于模糊 RBF 神经网络的无刷直流电机 PID 控制[J].计算机技术与发展,2022,32(08):15.[doi:10. 3969 / j. issn. 1673-629X. 2022. 08. 003]
　GU Wen-bin,YANG Sheng-sheng,WANG Xian-liang,et al.PID Control of Brushless DC Motor Based on Fuzzy RBF Neural Network[J].,2022,32(02):15.[doi:10. 3969 / j. issn. 1673-629X. 2022. 08. 003]

更新日期/Last Update: 1900-01-01