.png){kind=icon}
.png){kind=icon}
Return
Embodied intelligence machine tools: concepts, architecture and key technologies
.png){kind=icon}
PRE
AI
DOI：10.1007/s00170-026-18993-8.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
The intellectualization of machine tools constitutes a primary driving force for the advancement of the manufacturing industry. To enhance the robustness and generalizability of machine tools operating in dynamic and unstructured machining environments, and to achieve deep integration between intelligent algorithms and physical execution, this paper proposes the concept of an Embodied Intelligence Machine Tool (EIMT). Unlike conventional intelligent machine tools (IMTs), which rely primarily on disembodied intelligence decoupled from physical execution feedback, the EIMT paradigm emphasizes a closed-loop interactive system encompassing embodied perception, cognition, decision-making, and execution. A cloud-edge-device collaborative architecture grounded in the dual-process theory of cognition is established to enable machine tools to autonomously perceive, understand, reason about, and interact with the physical world. Four pivotal enabling technologies are systematically analyzed: embodied cognitive models, embodied simulation, Sim-to-Real transfer, and privacy and trust. A D-shaped profile milling case study is presented to verify the feasibility and effectiveness of the proposed framework. The EIMT concept is poised to advance machine tool evolution into a new era characterized by heightened autonomy, enhanced adaptability, and seamless human-machine-environment collaboration.
Keywords:
Intelligent manufacturing
Intelligent machine tool
Embodied intelligence
Cloud-edge-device collaboration
Dual-process theory
Journal
IF:
3.1
Papers:
1.8K
Citations:
6.4W