Research on Distribution Route Planning Model and Algorithm of Unmanned Aerial Vehicle (UAV) based on Improved Multi-objective Genetic Algorithm

Changjiang Gao, Yinan Chen, Xiaohai Tang

PROCEEDINGS OF 2021 2ND INTERNATIONAL CONFERENCE ON ARTIFICIAL INTELLIGENCE AND INFORMATION SYSTEMS (ICAIIS '21)(2021)

引用 3|浏览0
暂无评分
摘要
As an automatic and efficient delivery approach, the unmanned aerial vehicle (UAV) has received much attention in the field of electronic commerce and urban logistics for its natural ability to avoid traffic congestion. Regarding the arrangement of UAV delivery paths as a multi-objective path planning problem, we propose a novel genetic method based on the Non-Dominant Sorting Genetic Algorithm II (NSGA-II) with constraints to optimize the multi-UAV path planning problem, which is to minimize the total cost while maximizing customer satisfaction. In this algorithm, decision vectors are represented as 1-dimensional chromosomes, and two specialized variation operators are employed. These designs, as well as the properties of NSGA-II, enable us to find more non-crowding Pareto optimal solutions with less calculation. In a case study on part of the Solomon dataset, this algorithm found three sparsely distributed non-dominant solutions in one Pareto frontier within a relatively small population size and a short time.
更多
查看译文
关键词
Genetic Algorithm, Multi-objective Optimization, Path Planning, Unmanned Aerial Vehicle
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要