Abstract: Cuckoo search algorithm is a new nature-inspired optimization technique, which uses Lévy Flights random walk and Biased random walk to search new solutions iteratively. In Lévy Flights random walk, all individuals search new solutions around the best solution obtained so far. This may easy make the population converge to the best one. To avoid converging to the current optimal solution, this paper uses the reverse direction angle strategy to search new solutions probably, and proposes a cuckoo search algorithm with reverse direction angle and binomial crossover, called RBCS. In RBCS, a binomial crossover strategy is employed to enhance the search ability of Biased random walk. Compared with the standard cuckoo search algorithm, the experimental results show that the proposed strategies can improve the convergence speed and the solution quality of the algorithm for the continuous function optimization problems effectively. Compared with other improved cuckoo search algorithms and other evolutionary algorithms, the experimental results reveal that the proposed algorithm is competitive for the continuous function optimization problems.

Key words: cuckoo search algorithm, reverse direction angle, binomial crossover, function optimization problems

摘要： 布谷鸟搜索算法是一种新兴的自然仿生优化技术，其借用Lévy Flights随机走动和Biased随机走动搜索新的解。在Lévy Flights随机走动中，所有个体以当前种群获得的最优解为导向进行搜索，这容易导致种群趋同于该最优解。针对此问题，引入反方向视角使种群基于一定概率反向搜索，以避免趋同于当前最优解，并提出带反方向视角和二项式交叉的布谷鸟搜索算法。在提出的算法中，借用二项交叉操作以提高Biased随机走动的搜索能力。与标准的布谷鸟搜索算法对比，实验结果说明提出的策略能够有效地改善布谷鸟搜索算法求解连续函数优化问题的收敛速度和解的质量。与其他改进的布谷鸟搜索算法以及其他进化算法对比，实验结果说明提出的算法在求解连续函数优化问题上具有一定的竞争力。

关键词: 布谷鸟搜索算法, 反方向视角, 二项交叉, 函数优化问题