一种基于双策略协同的混凝土3D打印路径优化方法及系统
By employing a deep reinforcement learning method based on dual-strategy collaboration, the problems of discontinuity, frequent starts and stops, and long idle travel in concrete 3D printing path planning were solved, achieving high path continuity and low start and stop frequency, thus improving printing quality and efficiency.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- XI'AN UNIVERSITY OF ARCHITECTURE AND TECHNOLOGY
- Filing Date
- 2025-07-21
- Publication Date
- 2026-07-17
AI Technical Summary
Existing path planning methods for 3D concrete printing suffer from problems such as discontinuous paths, frequent starts and stops, long idle strokes, and sharp path turns, which affect printing efficiency and molding quality.
We employ a deep reinforcement learning approach based on dual-strategy collaboration. By constructing graph-structured data and adjacency matrices, we dynamically update path planning, introduce turning angle and start-stop penalty mechanisms, and design dual deep Q-network models to handle continuous path expansion and start-stop transitions respectively, generating G-code control instructions.
It achieves high continuity of the printing path, low start-stop frequency, and short idle stroke, improving printing quality and efficiency, and ensuring forming accuracy and structural stability.
Smart Images
Figure CN120911261B_ABST