Embodied Intelligence: A Literature Review

Han Ni; Longjiao Cui

doi:10.63313/JCSFT.9021

Authors

Han Ni University of Shanghai for Science and Technology, Shanghai, China Author
Longjiao Cui University of Shanghai for Science and Technology, Shanghai, China Author

DOI:

https://doi.org/10.63313/JCSFT.9021

Keywords:

Embodied intelligence, Multimodal large model, World model, Motion capture, Interaction design, Digital twin

Abstract

This paper systematically reviews the theoretical foundations, technological evolution, and current research landscape of embodied intelligence (Embodied AI), with particular attention to its interdisciplinary value and applications in the field of design. It first revisits key theories such as embodied cognition, in-telligence without representation, and morphological computation, illustrating how embodied intelligence promotes a paradigm shift in design logic through the coupling of “body–environment” interactions. It then outlines the de-velopment trajectory of technologies ranging from robotic hardware and mul-timodal fusion to large multimodal and world models, revealing their applica-tions across product, architectural, fashion, and service design. Furthermore, the study analyzes theoretical extensions and technical applications of embodied intelligence in user experience, interaction design, and sustainable design — including motion capture, multimodal interaction, and digital twins. Finally, it summarizes the technical, ethical, and interdisciplinary challenges currently faced and proposes future research directions: focusing on affective and light-weight intelligent agents, promoting paradigm innovation in “intelligent co-design,” and constructing a responsible interdisciplinary ecosystem.

References

[1] Bødker, S. (2020). Participatory design and embodied interaction revisited. Human–Computer Interaction, 35(2), 179–203.

[2] Google DeepMind. (2023). RT-2: Vision–language–action model for robotics. London, UK.

[3] IKEA Design Innovation Lab. (2024). PlaceAR+ digital twin for ergonomic product testing. Stockholm, Sweden.

[4] MIT Media Lab. (2024). Embodied intelligence in interactive design systems. Massachu-setts Institute of Technology.

[5] Norman, D. A. (2004). Emotional design: Why we love (or hate) everyday things. New York: Basic Books.

[6] Pengcheng Laboratory, & Sun Yat-sen University. (2024). Embodied intelligence and de-sign collaboration white paper. Shenzhen, China.

[7] Steelcase Research. (2022). Ergonomic optimization through pressure-sensing office chairs. Michigan, USA.

[8] Turing, A. M. (1950). Computing machinery and intelligence. Mind, 59(236), 433–460.

[9] Varela, F. J., Thompson, E., & Rosch, E. (1991). The embodied mind: Cognitive science and human experience. Cambridge, MA: MIT Press.

[10] Zhang, Y., Li, Q., & Chen, J. (2023). Embodied AI and multimodal perception integration. Robotics and Autonomous Systems, 168, 104312.

[11] Apple Human Interface Team. (2021). Magic Mouse and embodied gesture design evalua-tion. Cupertino, CA: Apple Inc.

[12] Brooks, R. A. (1991). Intelligence without representation. Artificial Intelligence, 47(1–3), 139–159.

[13] Clark, A., & Chalmers, D. (1998). The extended mind. Analysis, 58(1), 7–19.

[14] Dutch Design Week. (2023). Gravity-powered sustainable faucet design. Eindhoven, Neth-erlands.

[15] Fodor, J. A. (1975). The language of thought. Cambridge, MA: Harvard University Press.

[16] Gerstman, L. (2019). Design for all: Inclusive approaches in product usability. London: Routledge.

[17] Gibson, J. J. (1979). The ecological approach to visual perception. Boston: Houghton Mif-flin.

[18] Huawei Wearable Device Lab. (2023). Smartwatch adaptive recognition research report. Shenzhen, China.

[19] International Design for All Foundation. (2023). Adaptive silicone utensil project. Barcelo-na, Spain.

[20] Merleau-Ponty, M. (1962). Phenomenology of perception. London: Routledge.

[21] Microsoft Accessibility Team. (2024). Xbox adaptive controller usability report. Redmond, WA.

[22] MIT Media Lab. (2022). Embodied cognition in child-centered design research. Cambridge, MA.

[23] Norman, D. A. (2004). Emotional design: Why we love (or hate) everyday things. New York: Basic Books.

[24] Pfeifer, R., & Bongard, J. (2007). How the body shapes the way we think: A new view of in-telligence. Cambridge, MA: MIT Press.

[25] Pfeifer Bionic Design Lab. (2022). Bionic butterfly window for passive ventilation. Zurich, Switzerland.

[26] Samsung Design Lab. (2023). Embodied cognition evaluation framework for smart locks. Seoul, South Korea.

[27] Xiaomi User Experience Lab. (2023). User study of minimalist interaction in smart lamps. Beijing, China.

[28] iRobot Engineering Team. (2022). Roomba behavior-based navigation system report. Bedford, MA.

[29] Adobe Design Lab. (2022). Conversational AI for design collaboration using GPT-2. San Jose, CA.

[30] Autodesk Research. (2019). Applying 3D point cloud analysis in ergonomic design. To-ronto, Canada.

[31] Bulgari Design Studio. (2020). Precision robotic prototyping in jewelry design. Rome, Italy.

[32] Devlin, J., Chang, M.-W., Lee, K., & Toutanova, K. (2018). BERT: Pre-training of deep bidi-rectional transformers for language understanding. NAACL Proceedings, 4171–4186.

[33] DJI Engineering Team. (2024). Mini 3 Pro morphological computation structure report. Shenzhen, China.

[34] Franka Emika. (2018). Panda robot specifications and design report. Munich, Germany.

[35] Google DeepMind. (2021). Personalized reinforcement learning in smart lighting systems. London, UK.

[36] Google DeepMind. (2023). RT-2: Vision–language–action model for robotic reasoning. London, UK.

[37] Google Research. (2023). PaLM-E: An embodied multimodal large model for design tasks. Mountain View, CA.

[38] He, K., Gkioxari, G., Dollár, P., & Girshick, R. (2017). Mask R-CNN. IEEE Transactions on Pattern Analysis and Machine Intelligence, 42(5), 118–132.

[39] IKEA Research. (2020). Robot-assisted ergonomic testing of furniture handles. Malmö, Sweden.

[40] Lillicrap, T. P., Hunt, J. J., Pritzel, A., et al. (2016). Continuous control with deep reinforce-ment learning. ICLR Proceedings.

[41] Meta AI. (2023). I-JEPA: World model pretraining for physical reasoning. Menlo Park, CA.

[42] MIT Computer Science & Artificial Intelligence Lab. (2024). Real2Sim2Real precision mapping in home appliances. Cambridge, MA.