针对已有多任务优化算法寻优精度受限、计算时间成本过高等问题,提出一种基于改进灰狼算法的多任务优化算法(improved grey wolf algorithm based multitask optimization algorithm,IGWMTO)。该算法采用灰狼算法代替典型多任务算法中...针对已有多任务优化算法寻优精度受限、计算时间成本过高等问题,提出一种基于改进灰狼算法的多任务优化算法(improved grey wolf algorithm based multitask optimization algorithm,IGWMTO)。该算法采用灰狼算法代替典型多任务算法中的遗传算法,计算个体的因素等级和技能因子实现狼群分类,并以此更新个体隶属任务,引入扰动因子和动态权重改善狼群个体的更新方式。仿真测试结果表明:相比于传统多任务优化算法,所提算法在4个优化问题上的寻优精度的提升均超过了4.8%,计算耗时降低了70%以上。展开更多
The intelligent vehicle network uses advanced information technology to establish an efficient integrated vehicle transport system, which has received great attention in industry and academia, lnternet of Vehicles （...The intelligent vehicle network uses advanced information technology to establish an efficient integrated vehicle transport system, which has received great attention in industry and academia, lnternet of Vehicles （loV） in an urban environment is operated in a wireless environment with high bit error rate and interference. In addition, the wireless link between vehicles is likely to be lost. All of this makes it an important challenge to provide reliable mobile routing in an urban traffic environment. In this paper, a reliable routing algorithm with network coding （RR_ NC） is proposed to solve the above problems. A routing node sequence is discovered in IoV from source to destination by multi-metric ant colony optimization algorithm （MACO）, and then clusters are formed around every node in the sequence. By adding linear encoding into the transmission of data between vehicle＇s clusters, the RR_NC provides much more reliable transmission and can recover the original message in the event of disorder and loss of message. Simulations are taken under different scenarios, and the results prove that this novel algorithm can deliver the information more reliably between vehicles in real-time with lower data loss and communication overhead.展开更多
文摘针对已有多任务优化算法寻优精度受限、计算时间成本过高等问题,提出一种基于改进灰狼算法的多任务优化算法(improved grey wolf algorithm based multitask optimization algorithm,IGWMTO)。该算法采用灰狼算法代替典型多任务算法中的遗传算法,计算个体的因素等级和技能因子实现狼群分类,并以此更新个体隶属任务,引入扰动因子和动态权重改善狼群个体的更新方式。仿真测试结果表明:相比于传统多任务优化算法,所提算法在4个优化问题上的寻优精度的提升均超过了4.8%,计算耗时降低了70%以上。
基金supported by the Science and Technology Development Fund(No.037/2015/A1),Macao SAR,China
文摘The intelligent vehicle network uses advanced information technology to establish an efficient integrated vehicle transport system, which has received great attention in industry and academia, lnternet of Vehicles （loV） in an urban environment is operated in a wireless environment with high bit error rate and interference. In addition, the wireless link between vehicles is likely to be lost. All of this makes it an important challenge to provide reliable mobile routing in an urban traffic environment. In this paper, a reliable routing algorithm with network coding （RR_ NC） is proposed to solve the above problems. A routing node sequence is discovered in IoV from source to destination by multi-metric ant colony optimization algorithm （MACO）, and then clusters are formed around every node in the sequence. By adding linear encoding into the transmission of data between vehicle＇s clusters, the RR_NC provides much more reliable transmission and can recover the original message in the event of disorder and loss of message. Simulations are taken under different scenarios, and the results prove that this novel algorithm can deliver the information more reliably between vehicles in real-time with lower data loss and communication overhead.
