Multimodal multi-objective evolutionary algorithm for multiple path planning

•The differences between discrete and continuous MOPs are analyzed.•A special environmental selection strategy is proposed to maintain the diversity.•A diversity-based fitness indicator is proposed.•A novel multimodal multi-objective evolutionary algorithm is proposed. The multi-objective path plann...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:Computers & industrial engineering Ročník 169; s. 108145
Hlavní autoři: Yao, Xingyi, Li, Wenhua, Pan, Xiaogang, Wang, Rui
Médium: Journal Article
Jazyk:angličtina
Vydáno: Elsevier Ltd 01.07.2022
Témata:
Discrete optimization
Evolutionary algorithm
Multi-modal
Multi-objective optimization
Multiple Path planning
Multiple Path planning
Discrete optimization
Multi-objective optimization
Multi-modal
Evolutionary algorithm
ISSN:0360-8352, 1879-0550
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:•The differences between discrete and continuous MOPs are analyzed.•A special environmental selection strategy is proposed to maintain the diversity.•A diversity-based fitness indicator is proposed.•A novel multimodal multi-objective evolutionary algorithm is proposed. The multi-objective path planning problem has received much attention recently. Traditional solving methods try to find a single optimal path without considering the multiformity of the paths. In this study, we first analyze the situation that several different paths may have the same objective values, termed as multi-modal minimum path problems. To address these problems, we propose a novel solution-encoding method, which decreases the size of decision-space greatly. Then, to maintain the population diversity in the decision space, we propose an environmental selection strategy, in which the duplicate solutions are deleted first and then a second-selection method is adopted. Finally, an effective multi-objective evolutionary algorithm based on the special environmental selection is proposed, termed MMEA-SES. Through the experiments, the proposed method is proved effective and efficient compared to other state-of-the-art algorithms for multimodal multi-objective path planning.
ISSN:0360-8352
1879-0550
DOI:10.1016/j.cie.2022.108145