基于人工蜂鸟算法的机器人避障研究

Research on Robot Obstacle Avoidance Based on Artificial Hummingbird Algorithm

针对传统避障算法搜索空间复杂、易陷入局部极值、避障效率低下等问题,提出了一种融合多种改进策略的人工蜂鸟算法。首先,利用Liebovitch映射生成初始候选解,从而提升人工蜂鸟种群的丰富性;其次,利用群体上个体间的差异变异对个体进行扰动,从而保留优质个体,引导搜索过程逼近全局最优解,避免过早收敛;然后,在引导觅食阶段引入黄金正弦因子,有利于缩小蜂鸟种群搜索范围,提高收敛精度和速度;最后,进行机器人避障仿真实验。实验结果表明,IAHA算法各项指标均优于其他算法,具有更好的优化性能,同时提高了机器人避障效率。

To solve the problems of complex search space,tend to fall into local extremum and low obstacle avoidance efficiency of traditional obstacle avoidance algorithms,a improved artificial hummingbird algorithm(IAHA)is proposed.Firstly,Liebovitch mapping is used to generate the initial candidate solution,so as to improve the richness of artificial hummingbird population.Secondly,the difference and variation among individuals on the population are used to disturb individuals,so as to retain high-quality individuals,search towards the global optimal solution,and avoid premature convergence.Then,the golden sine factor is introduced in the guided foraging stage,which is conducive to reducing the search range of hummingbirds and improving the convergence accuracy and speed.Finally,the simulation experiment of robot obstacle avoidance is performed.The experiment results show that the IAHA algorithm has better optimization performance than other algorithms,and improves the robot obstacle avoidance efficiency.