Landscape design data processing method based on artificial intelligence
By collecting and fusing multi-source heterogeneous data, extracting design constraints driven by knowledge graphs, and dynamically adjusting the weights of reinforcement learning models, combined with physical model-based adversarial design, the problem of insufficient design efficiency and quality in multi-source heterogeneous data processing in existing technologies has been solved, thus realizing the scientific and rational nature of landscape design.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- GUANGDONG OCEAN UNIVERSITY
- Filing Date
- 2025-07-02
- Publication Date
- 2026-05-29
AI Technical Summary
Existing landscape design data processing methods struggle to construct spatiotemporally correlated multimodal datasets when faced with multi-source heterogeneous data. They also lack the ability to dynamically adjust the weights of design requirement indicators, resulting in insufficient physical feasibility of design schemes and unscientific multi-objective optimization and scheme verification. Consequently, design efficiency and quality need to be improved.
By collecting and fusing multi-source heterogeneous data, extracting design constraints driven by knowledge graphs, dynamically adjusting weights using reinforcement learning models, generating adversarial designs by coupling physical models, and employing multi-objective optimization and virtual-real fusion verification, the scientific and rational design scheme is achieved.
Accurately identify hard constraints such as terrain and ecology, as well as cultural attributes, and dynamically adjust the weights of factors such as ecology and cost to ensure the physical feasibility and scientific validity of the design scheme, thereby improving design efficiency and quality.
Smart Images
Figure CN120781685B_ABST