Moth-flame optimization(MFO)is a novel metaheuristic algorithm inspired by the characteristics of a moth’s navigation method in nature called transverse orientation.Like other metaheuristic algorithms,it is easy to f...Moth-flame optimization(MFO)is a novel metaheuristic algorithm inspired by the characteristics of a moth’s navigation method in nature called transverse orientation.Like other metaheuristic algorithms,it is easy to fall into local optimum and leads to slow convergence speed.The chaotic map is one of the best methods to improve exploration and exploitation of the metaheuristic algorithms.In the present study,we propose a chaos-enhanced MFO(CMFO)by incorporating chaos maps into the MFO algorithm to enhance its performance.The chaotic map is utilized to initialize the moths’population,handle the boundary overstepping,and tune the distance parameter.The CMFO is benchmarked on three groups of benchmark functions to find out the most efficient one.The performance of the CMFO is also verified by using two real engineering problems.The statistical results clearly demonstrate that the appropriate chaotic map(singer map)embedded in the appropriate component of MFO can significantly improve the performance of MFO.展开更多
This paper presents an Enhanced Moth-Flame Optimization （EMFO） technique based on Cultural Learning （CL） and Gaussian Mutation （GM）. The mechanism of CL and the operator of GM are incorporated to the original al...This paper presents an Enhanced Moth-Flame Optimization （EMFO） technique based on Cultural Learning （CL） and Gaussian Mutation （GM）. The mechanism of CL and the operator of GM are incorporated to the original algorithm of Moth-Flame Optimization （MFO）. CL plays an important role in the inheritance of historical experiences and stimulates moths to obtain information from flames more effectively, which helps MFO enhance its searching ability. Furthermore, in order to overcome the disadvantage of trapping into local optima, the operator of GM is introduced to MFO. This operator acts on the best flame in order to generate several variant ones, which can increase the diversity. The proposed algorithm of EMFO has been comprehensively evaluated on 13 benchmark functions, in comparison with MFO. Simulation results verify that EMFO shows a significant improvement on MFO, in terms of solution quality and algorithmic reliability.展开更多
基金supported by the Military Science Project of the National Social Science Foundation of China(15GJ003-141)
文摘Moth-flame optimization(MFO)is a novel metaheuristic algorithm inspired by the characteristics of a moth’s navigation method in nature called transverse orientation.Like other metaheuristic algorithms,it is easy to fall into local optimum and leads to slow convergence speed.The chaotic map is one of the best methods to improve exploration and exploitation of the metaheuristic algorithms.In the present study,we propose a chaos-enhanced MFO(CMFO)by incorporating chaos maps into the MFO algorithm to enhance its performance.The chaotic map is utilized to initialize the moths’population,handle the boundary overstepping,and tune the distance parameter.The CMFO is benchmarked on three groups of benchmark functions to find out the most efficient one.The performance of the CMFO is also verified by using two real engineering problems.The statistical results clearly demonstrate that the appropriate chaotic map(singer map)embedded in the appropriate component of MFO can significantly improve the performance of MFO.
基金The work is supported by National Natural Science Foundation of China （Grant No. 51707069）, the State Key Laboratory of Alternate Electrical Power System with Renewable Energy Sources （Grant No. LAPS 18001）, National Natural Science Foundation of China （Grant No. 51277080）, MOE Key Laboratory of Image Processing and Intelligence Control, Wuhan, China （Grant No. IPIC2015-01）, and State Key Program of National Natural Science Foundation of China （Grant No.51537003）.
文摘This paper presents an Enhanced Moth-Flame Optimization （EMFO） technique based on Cultural Learning （CL） and Gaussian Mutation （GM）. The mechanism of CL and the operator of GM are incorporated to the original algorithm of Moth-Flame Optimization （MFO）. CL plays an important role in the inheritance of historical experiences and stimulates moths to obtain information from flames more effectively, which helps MFO enhance its searching ability. Furthermore, in order to overcome the disadvantage of trapping into local optima, the operator of GM is introduced to MFO. This operator acts on the best flame in order to generate several variant ones, which can increase the diversity. The proposed algorithm of EMFO has been comprehensively evaluated on 13 benchmark functions, in comparison with MFO. Simulation results verify that EMFO shows a significant improvement on MFO, in terms of solution quality and algorithmic reliability.
文摘针对无人机长期跟踪过程中尺度变换导致目标丢失和跟踪精度低的问题,提出了一种基于飞蛾扑火优化(moth-flame optimization,MFO)的尺度比例感知空间长期跟踪器。首先,设计了高斯初始化以代替飞蛾扑火优化算法的随机初始化策略,降低优化算法在跟踪过程中的计算复杂度,减少算力浪费;其次,结合快速梯度直方图特征,构建了改进的飞蛾扑火优化跟踪器;然后,为了解决无人机航拍长期跟踪中目标尺度变化的问题,设计了一种自适应尺度变换的判别尺度空间跟踪(discriminative scale space tracking,DSST)算法,进一步提出了一种尺度比例感知空间跟踪器,解决了尺度滤波器中因长宽比固定而导致的跟踪漂移;同时,分析了滤波器响应峰值在各背景下的变化情况,提出了一种能反映环境变化下跟踪置信度的指标,并通过置信度将MFO优化跟踪框架与尺度比例感知空间跟踪器相结合,解决了尺度变化与长期跟踪目标丢失的问题;最后,在无人机长期跟踪数据集上开展了性能验证。结果表明:提出的算法可有效防止漂移现象的发生,提升跟踪效率;与目前跟踪领域中12种同类文献算法进行对比可知,提出的算法精度较高,满足实时性,能够有效解决无人机长期跟踪下的尺度变化及目标丢失等问题。
