Animal Collective Behavior: Mechanisms and Applications
Xiaofeng Liu,Guo Chen,Yu Liu,Xi Wang
DOI: https://doi.org/10.1360/tb-2023-0165
2023-01-01
Abstract:Collective behavior is a complex and fascinating phenomenon that has intrigued researchers for decades. The study of collective behavior encompasses various aspects such as information transfer, decision-making processes, locomotion, and group synchronization. In the past decades, researchers have investigated the mechanisms and bionic applications of animal collective behavior, which have inspired and contributed to various fields such as biology, psychology, sociology, computer science, and robotics. In this review, we aim to: (1) Explore the theoretical models and mechanisms underlying swarm behavior from the perspectives of collective decision-making, motor control and neural basis; (2) examine how these mechanisms are applied to computer science and robotics; and (3) suggest how new techniques can be used to further investigate the potential directions of animal collective behavior. This review proposes that collective decision-making originates from the collective perception of animals. Individuals utilize sensory or chemical information to perceive their environment and establish communication with their peers. Through interactions among individuals, a consensus is gradually reached at the group level. During the process of perceiving the environment and engaging in internal communication, individual differences play a role and exert influence on collective decision-making to some extent. The presence of complex survival needs, environmental constraints, and physical limitations within animal populations can lead to the emergence of diverse movement structures, which serve as manifestations of collective behavior. Beneath the phenomena of information interactions and movement structures, there exist neural mechanisms that regulate animal behavior. The research scope of collective behavior encompasses the study of individuals and groups. At the individual level, we can investigate how heterogeneous individuals impact collective decision-making and how neural mechanisms within individuals govern the regulation of behavior related to collective behavior and social learning. Collective decision-making focuses on the transmission of information and coordination among individual animals. The mutual coordination among individual animals gives rise to specific motor control methods, and different motor control methods contribute to distinct collective decision modes. The study of animal neural mechanisms provides an explanation of how animals regulate their behavior to engage in collective decision-making and adopt different motor control styles. Concurrently, collective decision-making and the ongoing influence of motor control can provide feedback and shape the unique neural circuits of animals. By exploring these interconnected aspects, we can enhance our understanding of the intricate dynamics of collective behavior and the underlying neural mechanisms that govern it. This review also explores various biomimetic applications derived from collective behavior research, including the design of biomimetic optimization algorithms, multi-agent cooperative algorithms, and the application of bionic swarm robots. Furthermore, the review provides valuable insights for the future development of swarm behavior research. It suggests that the study of collective behavior can leverage advanced equipment and data analysis techniques to delve into more intricate mechanisms. The progress of research is also reliant on interdisciplinary collaboration among researchers. Additionally, fostering an open data analysis and processing platform can facilitate further advancements in the field. In conclusion, the study of animal collective behavior is a complex and multidimensional discipline with the potential to bring substantial benefits to various areas of research.