Abstract: Teaching-Learning-Based Optimization (TLBO) algorithm is a powerful evolutionary algorithm that has recently been widely applied in a variety of optimization problems.However,in the later period of evolution of the TLBO algorithm,the diversity of learners will be degraded with the increasing iteration of evolution and the smaller scope of solutions,which leads to trap in local optima and premature convergence.For this reason,an improved version of TLBO algorithm based on Laplace distribution and Balwin learning effect (LBTLBO) was presented.Laplace distribution was used to expand exploration space,and the Balwin learning effect was applied to make good use of experience information to identify more promising solutions to make the algorithm more competitive.The experimental performances verified that LBTLBO algorithm enhanced the solution accuracy and quality compared to original TLBO,TLBO algorithm based on Laplace distribution (LTLBO),TLBO algorithm based on Balwin learning effect (BTLBO) as well as various versions of TLBO.LBTLBO was used to solve Unmanned Aerial Vehicle(UVA)path planning problem.The simulation experiments showed that the improved algorithm could get more accurate path and convergence rate in the UVA path planning problem.

Key words: teaching-learning-based optimization, Laplace distribution, Balwin learning effect, unmanned aerial vehicle path planning

摘要： 教与学优化算法(TLBO)是一种进化能力非常强大的算法,近年被广泛应用于各种优化问题。但是,在TLBO算法进化的后期,随着进化迭代次数的增加和求解范围的缩小,种群的多样性逐渐降低,从而导致陷入局部最优和过早收敛。基于此,提出一种基于拉普拉斯分布和鲍德温学习效应的教与学优化算法(LBTLBO)。该算法利用拉普拉斯分布的扰动来拓展探索空间,采用鲍德温学习效应识别出更多有前途的解,使算法更具有竞争性。实验结果表明,与原始TLBO、基于拉普拉斯分布的TLBO、基于鲍德温学习的TLBO以及改进版本的TLBO进行比较,LBTLBO提高了解的精度,具有很强的竞争力。最后,将LBTLBO应用于无人机航路规划问题,并进行了仿真实验,结果显示,与上述改进版本的TLBO相比,LBTLBO能获得更加准确的路径与收敛速度。

关键词: 教与学优化算法, 拉普拉斯分布, 鲍德温学习效应, 无人机航路规划