面向车载AI知识库的数据处理方法
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- NANJING PENGCHEN QIZHI TECH CO LTD
- Filing Date
- 2026-06-18
- Publication Date
- 2026-07-17
AI Technical Summary
The existing data processing system of in-vehicle AI knowledge base suffers from outdated data and delayed updates, which cannot meet the high real-time, high precision and high reliability requirements of intelligent connected vehicles. It also cannot accurately distinguish the degree of failure of knowledge data, resulting in problems such as interaction errors, driving assistance judgment errors and false alarms in fault detection.
By adopting a standardized model for collecting knowledge metadata and a standardized calculation model for measuring knowledge obsolescence and update lag, the obsolescence and update lag of knowledge are calculated through mathematical expressions. Combined with hierarchical judgment criteria, this enables accurate quantification and automated updating of knowledge.
This has improved the data processing accuracy of the in-vehicle AI knowledge base, eliminated expired and invalid data, reduced maintenance costs, improved the timeliness and accuracy of the knowledge base, ensured the stability and reliability of the in-vehicle AI system, and avoided system lag and response delays.
Smart Images
Figure CN122412432A_ABST