year 13, Issue 1 (5-2026)
CJS 2026, 13(1): 94-108 |
Back to browse issues page
Download citation:
BibTeX | RIS | EndNote | Medlars | ProCite | Reference Manager | RefWorks
Send citation to:
BibTeX | RIS | EndNote | Medlars | ProCite | Reference Manager | RefWorks
Send citation to:
Mohammadi Kouhbanani A, Ahmadinia M. The Evolution of Swarm Robotics in the Context of Nature Inspired Computing: A Scientometric Study. CJS 2026; 13 (1) :94-108
URL: http://cjs.mubabol.ac.ir/article-1-435-en.html
URL: http://cjs.mubabol.ac.ir/article-1-435-en.html
Department of Computer Engineering, Ke.C., Islamic Azad University, Kerman, Iran , aminmohamadi.info@gmail.com
Abstract: (11 Views)
Background and aim: Swarm robotics and nature-inspired computing, as interdisciplinary fields, have grown considerably in recent years by integrating artificial intelligence, computational biology, optimization algorithms, and autonomous robotic systems. Despite the rapid increase in scientific publications, a comprehensive picture of the intellectual evolution, knowledge structure, and emerging trends in this field has not yet been presented. Therefore, the present study was conducted with the aim of performing a scientometric analysis to identify research trends, knowledge structures, influential scientific networks, and emerging research gaps in swarm robotics and nature-inspired computing.
Materials and methods: This research is of an applied type with a descriptive-analytical approach and is based on the scientometric method. The research population included all documents indexed in Scopus in the field of swarm robotics and nature-inspired computing during the years 2000 to 2025. Bibliographic data were extracted and analyzed using VOSviewer software. In order to map the scientific structure of this field, keyword co-occurrence, co-authorship, co-citation, and bibliographic coupling networks were examined. Furthermore, in addition to scientific performance analysis, citation impact indicators, including the average number of citations per article, were analyzed to evaluate the trend of research quality evolution over time.
Findings: The quantitative findings of the study revealed that scientific publications in this field followed an upward trend from 2000 to 2025, accelerating significantly after 2020. Pearson correlation analysis confirmed a positive, strong, and statistically significant relationship between the annual number of publications and both the average citations per paper and the annual h-index, indicating a simultaneous growth in the quantity and quality of research output. Among the scientific journals, Swarm Intelligence secured the highest level of scholarly authority, receiving 2,983 citations from 55 publications. Geographically, China achieved the highest quantitative share with 225 publications, while Belgium recorded the highest academic impact with 4,164 citations. Iran ranked among the top 20 leading nations, accumulating 540 citations from 17 publications. Furthermore, co-occurrence analysis identified the term “swarm intelligence” as the core concept of the network, with 716 occurrences and a total link strength of 4,176. It also revealed a research paradigm shift since 2022, moving away from traditional simulation models toward “reinforcement learning” and “autonomous agents.”
Conclusion: The scientific growth of swarm robotics has been accelerated, and its effectiveness (citation rate) has a direct correlation with international collaborations and the interdisciplinary nature. However, the lack of a significant correlation between the development of theoretical algorithms and practical implementations was identified as the main structural gap in this field. Accordingly, the future direction of research should be oriented from abstract single-agent models toward the integration of embodied artificial intelligence and reinforcement learning in real robots to enable the transition from theory to practice.
Materials and methods: This research is of an applied type with a descriptive-analytical approach and is based on the scientometric method. The research population included all documents indexed in Scopus in the field of swarm robotics and nature-inspired computing during the years 2000 to 2025. Bibliographic data were extracted and analyzed using VOSviewer software. In order to map the scientific structure of this field, keyword co-occurrence, co-authorship, co-citation, and bibliographic coupling networks were examined. Furthermore, in addition to scientific performance analysis, citation impact indicators, including the average number of citations per article, were analyzed to evaluate the trend of research quality evolution over time.
Findings: The quantitative findings of the study revealed that scientific publications in this field followed an upward trend from 2000 to 2025, accelerating significantly after 2020. Pearson correlation analysis confirmed a positive, strong, and statistically significant relationship between the annual number of publications and both the average citations per paper and the annual h-index, indicating a simultaneous growth in the quantity and quality of research output. Among the scientific journals, Swarm Intelligence secured the highest level of scholarly authority, receiving 2,983 citations from 55 publications. Geographically, China achieved the highest quantitative share with 225 publications, while Belgium recorded the highest academic impact with 4,164 citations. Iran ranked among the top 20 leading nations, accumulating 540 citations from 17 publications. Furthermore, co-occurrence analysis identified the term “swarm intelligence” as the core concept of the network, with 716 occurrences and a total link strength of 4,176. It also revealed a research paradigm shift since 2022, moving away from traditional simulation models toward “reinforcement learning” and “autonomous agents.”
Conclusion: The scientific growth of swarm robotics has been accelerated, and its effectiveness (citation rate) has a direct correlation with international collaborations and the interdisciplinary nature. However, the lack of a significant correlation between the development of theoretical algorithms and practical implementations was identified as the main structural gap in this field. Accordingly, the future direction of research should be oriented from abstract single-agent models toward the integration of embodied artificial intelligence and reinforcement learning in real robots to enable the transition from theory to practice.
Type of Study: Orginal |
Subject:
Scientometrics
Received: 2026/05/21 | Revised: 2026/07/25 | Accepted: 2026/08/5 | ePublished: 2026/08/18
Received: 2026/05/21 | Revised: 2026/07/25 | Accepted: 2026/08/5 | ePublished: 2026/08/18
Send email to the article author
| Rights and permissions | |
 |
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License. |
