Embodied Intelligent Robots: Technological Evolution, Core Systems, Challenges, and Future Outlook
DOI: https://doi.org/10.62517/jike.202604117
Author(s)
Xin Yuan
Affiliation(s)
Guangzhou Huali College, Guangzhou, Guangdong, China
Abstract
As an innovative product of multi-disciplinary integration, embodied intelligent robots break the functional limitations of traditional robots. By achieving in-depth interaction with the physical environment to realize autonomous decision-making, they have become a core support for intelligent transformation. This paper systematically sorts out their four major development stages, constructs the "perception-decision-action-feedback-learning" core technical system, and details key technologies such as multi-modal perception and "big brain-small brain" collaborative decision-making. Meanwhile, it analyzes four major current challenges: lack of standardization, dependence on core components, talent shortage, and ethical risks, and proposes five development directions: technological innovation, industrial collaboration, scene penetration, policy improvement, and human-robot symbiosis. The research provides multi-dimensional references for scientific research, industrial application, and policy formulation of embodied intelligent robots, helping them upgrade towards general intelligence and promoting the construction of a new production relationship of "technology empowerment and human leadership".
Keywords
Embodied Intelligent Robots; Embodied Intelligence; Disembodied Intelligence; General Intelligence; Human-Robot Symbiosis
References
[1] TURING A M. Computing machinery and intelligence[J]. Mind, 1950, LIX(236): 433-460.
[2] LIU Huaping, GUO Di, SUN Fuchun, et al. Morphology-based embodied intelligence: Historical retrospect and research progress[J]. Acta Automatica Sinica, 2023, 49(6): 1131−1154.
[3] GIBSON J J. The ecological approach to visual perception: Classic edition[M]. Psychology Press, 1979.
[4] VARELA F J, THOMPSON E, ROSCH E. The embodied mind: Cognitive science and human experience[M]. The MIT Press, 1991.
[5] BROOKS R. A robust layered control system for a mobile robot[J]. IEEE Journal on Robotics and Automation, 1986, 2(1): 14-23.
[6] PFEIFER R. Understanding intelligence[M]. Cambridge, Massachusetts: MIT Press, 1999.
[7] COLLINS S, RUINA A, TEDRAKE R, et al. Efficient bipedal robots based on passive-dynamic walkers[J]. Science, 2005, 307(5712): 1082-1085.
[8] PFEIFER R. How the body shapes the way we think: A new view of intelligence[M]. Cambridge: MIT Press, 2006.
[9] LEVINE S, FINN C, DARRELL T, et al. End-to-end training of deep visuomotor policies[J]. Journal of Machine Learning Research, 2016, 17(39): 1-40.
[10] HAARNOJA T, ZHOU A, HARTIKAINEN K, et al. Soft actor-critic algorithms and applications[EB/OL]. (2023-04-18). http://arxiv.org/abs/1812.05905.
[11] OpenAI, ANDRYCHOWICZ M, BAKER B, et al. Learning dexterous in-hand manipulation[EB/OL]. (2025-02-14). http://arxiv.org/abs/1808.00177.
[12] ZITKOVICH B, YU T, XU S, et al. RT-2: Vision-language-action models transfer web knowledge to robotic control[C]//Proceedings of The 7th Conference on Robot Learning. PMLR, 2023: 2165-2183.
[13] SCHÖLKOPF B, LOCATELLO F, BAUER S, et al. Towards causal representation learning[EB/OL]. (2025-02-14)[2025-02-14]. http://arxiv.org/abs/2102.11107.
[14] BASU S, TATLIDIL S, KIM M H, et al. Invited: Using causal information to enable more efficient robot operation[C]//Proceedings of the 61st ACM/IEEE Design Automation Conference. New York, NY, USA: Association for Computing Machinery, 2024: 1-4.
[15] YANG Z, WANG T, LIN Y, et al. A vision chip with complementary pathways for open-world sensing[J]. Nature, 2024, 629(8014): 1027-1033.
[16] WEI Shaomeng, ZHANG Jiangmin, SHI Hui, et al. Remaining useful life prediction of relative density high-dimensional kernel estimation for systems based on multi-sensor data fusion[J]. Computer Integrated Manufacturing System, 2024, 30(12): 4493-4507.
