基于增量式自组织的小脑学习模型及其应用方法
By combining incremental self-organizing networks and neuronal regeneration mechanisms, the problem of insufficient adaptability and learning ability of mobile robots in behavioral decision-making in unknown environments is solved, and rapid and stable decision-making ability is achieved.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- GANTRY LAB
- Filing Date
- 2023-10-12
- Publication Date
- 2026-07-17
AI Technical Summary
Existing mobile robot behavior decision-making methods have poor adaptability in unknown environments. The fixed network topology of traditional cerebellar models limits their continuous learning ability, resulting in weak behavior decision-making ability in changing environments. Furthermore, existing learning algorithms suffer from data efficiency issues and limited memory capacity.
We designed a cerebellar learning model based on an incremental self-organizing network, combined with a neuronal regeneration mechanism. Through interactive learning via exploration and practice, we achieved incremental learning of agent behavior decisions in unknown environments. By employing an incremental self-organizing developmental network and a neuronal regeneration mechanism, we enhanced learning ability and environmental adaptability.
It improves the speed and stability of mobile robots' behavioral decision-making in unknown environments, solves the problems of fixed memory capacity and insufficient learning ability in traditional models, and realizes dynamic real-time adjustment and incremental knowledge storage.
Smart Images
Figure CN117422124B_ABST