TY - JOUR
T1 - Cognitive navigation
AU - Yang, Yi
AU - Wang, Tao
AU - Pan, Miaoxin
AU - Zhang, Yiqing
AU - Yue, Yufeng
AU - Fu, Mengyin
N1 - Publisher Copyright:
© Science China Press 2026.
PY - 2026/5
Y1 - 2026/5
N2 - Currently, new-generation artificial intelligence technologies, such as foundation models and embodied intelligence, are developing rapidly, significantly expanding the connotation and extension of navigation technology. Cognitive navigation, by interacting with the environment, achieves learning, growth, and evolution. It represents a new navigation paradigm formed by the deep integration of cognitive science, navigation technology, artificial intelligence, and real-world application scenarios. The core of cognitive navigation lies in the growth and evolution of cognitive abilities under spatiotemporal coupling, with its technological essence reflected in the four dimensions of “ability-interpretability-generalizability-evolvability”, forming an adaptive navigation architecture based on the ternary environment-task-platform coupling. First, this review summarizes the history and current state of navigation technology, explores the foundations of cognitive navigation in the fields of psychology, neuroscience, and artificial intelligence, and emphasizes the need to advance navigation technology into the cognitive space. Then it provides a detailed introduction to the framework of cognitive navigation along with the latest advances in perception, decision, and execution. Finally, the review explores the cognitive navigation paradigm based on the mechanism “ability-interpretability-generalizability-evolvability”, emphasizing the need for further research on next-generation cognitive navigation technologies that integrate stable perception, intelligent decision, and adaptive execution, with the aim of achieving high efficiency, strong versatility, and self-evolving capabilities for “one brain, multiple forms” and “one machine, multiple uses”.
AB - Currently, new-generation artificial intelligence technologies, such as foundation models and embodied intelligence, are developing rapidly, significantly expanding the connotation and extension of navigation technology. Cognitive navigation, by interacting with the environment, achieves learning, growth, and evolution. It represents a new navigation paradigm formed by the deep integration of cognitive science, navigation technology, artificial intelligence, and real-world application scenarios. The core of cognitive navigation lies in the growth and evolution of cognitive abilities under spatiotemporal coupling, with its technological essence reflected in the four dimensions of “ability-interpretability-generalizability-evolvability”, forming an adaptive navigation architecture based on the ternary environment-task-platform coupling. First, this review summarizes the history and current state of navigation technology, explores the foundations of cognitive navigation in the fields of psychology, neuroscience, and artificial intelligence, and emphasizes the need to advance navigation technology into the cognitive space. Then it provides a detailed introduction to the framework of cognitive navigation along with the latest advances in perception, decision, and execution. Finally, the review explores the cognitive navigation paradigm based on the mechanism “ability-interpretability-generalizability-evolvability”, emphasizing the need for further research on next-generation cognitive navigation technologies that integrate stable perception, intelligent decision, and adaptive execution, with the aim of achieving high efficiency, strong versatility, and self-evolving capabilities for “one brain, multiple forms” and “one machine, multiple uses”.
KW - artificial intelligence
KW - cognitive evolution mechanism
KW - cognitive navigation
KW - embodied cognition
KW - navigation system
UR - https://www.scopus.com/pages/publications/105037455237
U2 - 10.1007/s11432-025-4852-4
DO - 10.1007/s11432-025-4852-4
M3 - Review article
AN - SCOPUS:105037455237
SN - 1674-733X
VL - 69
JO - Science China Information Sciences
JF - Science China Information Sciences
IS - 5
M1 - 151201
ER -