互联网 qkzz.net
全刊杂志网:首页 > 女性 > 文章正文
刊社推荐

基于当代学习离散粒子群算法的多机器人任务分配


□ 余伶俐 蔡自兴

   (中南大学 信息科学与工程学院 长沙 410083)
  
  摘 要:针对多机器人协同控制中的任务分配问题,首先综合考虑机器人完成任务的效率、机器人自身能力以及任务本身性质各因素,建立了多机器人任务分配的数学模型。而后提出一种基于当代学习机制的离散粒子群算法进行高效求解,该算法设计了准确的粒子运动方程,并加入扰动算子保持粒子多样性,使其迅速跳出局部最优,增加了算法空间探索能力。实验结果表明:在小规模任务数情况下,算法能精确寻到最优,稳定性表现极佳且优于现有算法。在中大规模任务数情况下算法也表现出强寻优能力,实验验证了模型的合理性和算法的优越性。
  关键词:多机器人; 离散粒子群; 任务分配; 扰动因子
  中图分类号:TP242文献标志码:A
  文章编号:1001-3695(2009)05-1691-04
  
  Multirobot mission assignment based on current learning
  discrete particle swarm optimization algorithm
  YU Lingli CAI Zixing
  (School of Information Science & Engineering Central South University Changsha 410083 China)
  Abstract:Multirobot mission assignment mathematical model was established firstly which considered three factors comprehensively: executing mission efficiency robot ability and mission properties. This paper proposed current learning discrete particle swarm optimization algorithm(CLDPSO) to solve multirobot mission assignment with highly efficiently. The algorithm designed an exact particles kinetic equation. When decreased algorithm diversity to a certain threshold,added a perturbation operator to jump out local optimum quickly and to improve the search ability. The experiment results show that CLDPSO can reach the best result and its stability is the best among existing algorithms when the number of missions is small scale. When the number of missions is middle or large scale the searching optimization ability is also strong. Those experiments prove that the model is reasonably and CLDPSO algorithm is the advantage. ......
很抱歉,暂无全文,若需要阅读全文或喜欢本刊物请联系《计算机应用研究》杂志社购买。
欢迎作者提供全文,请点击编辑
分享:
 

了解更多资讯,请关注“木兰百花园”
分享:
 
精彩图文


关键字
支持中国杂志产业发展,请购买、订阅纸质杂志,欢迎杂志社提供过刊、样刊及电子版。
关于我们 | 网站声明 | 刊社管理 | 网站地图 | 联系方式 | 中图分类法 | RSS 2.0订阅 | IP查询
全刊杂志赏析网 2017