摘要
为缩短农业机器人全局路径规划的长度和时间并获得更好的安全路径,提出了一种采用多策略改进黑猩猩算法(MIChOA)的路径规划方法。首先,对传统的黑猩猩算法(ChOA)进行改进,在种群初始化阶段引入佳点集策略提高种群的多样性;其次,根据黑猩猩实际寻优过程提出正态随机余弦收敛因子策略,平衡了算法全局勘探与局部开发能力;然后,在算法陷入局部最优停滞时采用贪婪竞争机制的停滞扰动策略,加快算法跳出局部最优并准确找到全局最优解;最后,利用标准测试函数验证MIChOA算法的寻优性能,建立了具有111个障碍物的环境栅格地图开展仿真实验,将MIChOA算法应用于农业机器人路径规划,并与其他4种较为优秀的改进ChOA算法进行比较。结果表明:MIChOA算法在单峰和复杂多峰函数上均具有较高的寻优精度、稳定的收敛性和良好的鲁棒性;MIChOA算法的路径搜索性能优于其他4种改进ChOA算法,其中路径长度缩短了28.01%,寻优时间和迭代次数分别减少了54.58%和85.87%。
To shorten the length and time of global path planning for agricultural robots and to obtain better safe paths,a multistrategy improved chimp optimization algorithm(MIChOA)is proposed.Firstly,a good point set strategy is introduced in the population initialisation stage to improve the population diversity,thus improving the traditional chimp optimization algorithm;then,a normal stochastic cosine convergence factor strategy is proposed based on the actual chimp search process to balance the global exploration and local exploitation capability of the algorithm;secondly,a stagnation perturbation strategy with a greedy competition mechanism is used when the algorithm falls into a local optimum stagnation to speed up the algorithm’s jumping out of the local optimum and finding the global optimum.Finally,a standard test function is used to verify the performance of the MIChOA algorithm in finding the global optimum,and an environmental raster map with 111 obstacles is created for corresponding experimental simulations.The MIChOA algorithm is applied to agricultural robot path planning and compared with the four relatively excellent improved ChOA algorithms.The results show that the MIChOA algorithm features high search accuracy,stable convergence and good robustness in terms of both singlepeak and complex multipeak functions;the MIChOA algorithm outperforms the other four improved ChOA algorithms in path search performance,with a 28.01%reduction in path length,54.58%reduction in search time and 85.87%reduction in the number of iterations.
作者
艾尔肯·亥木都拉
穆占海
郑威强
Aierken HAIMUDULA;MU Zhanhai;ZHENG Weiqiang(School of Smart Manufacturing Modern Industrial,Xinjiang University,Urumqi 830047,China)
出处
《西安交通大学学报》
EI
CAS
CSCD
北大核心
2023年第8期161-171,共11页
Journal of Xi'an Jiaotong University
基金
国家自然科学基金资助项目(52265039)。
关键词
农业机器人
黑猩猩优化算法
路径规划
局部最优
agricultural robot
chimp optimization algorithm
path planning
local optimum
