一种垃圾回收方法
By combining multimodal recognition and real-time capacity monitoring with user category information, the robot achieves accurate classification and dynamic path planning for waste recycling, solving the problems of inaccurate identification and insufficient path scheduling in existing technologies, and improving the accuracy, safety and efficiency of waste recycling.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- GUANGZHOU YINGSHI TECHNOLOGY CO LTD
- Filing Date
- 2025-10-24
- Publication Date
- 2026-07-17
AI Technical Summary
Existing waste recycling robots struggle to accurately acquire the physical characteristics inside bags when the bags are intact. They also lack integration with user category information, making it impossible to achieve dynamic path planning, priority enhancement for hazardous materials, and closed-loop disposal. This results in insufficient classification accuracy, operational continuity, and system safety.
A multimodal recognition module is used to obtain the physical characteristics of garbage bags, and classification decisions are made in combination with user category information. The path is updated in real time through a capacity monitoring module, and dynamic operation route optimization is achieved by configuring hazardous materials isolation boxes and navigation planning modules, forming a closed-loop process.
It improves the accuracy and interpretability of entrance identification, realizes hierarchical diversion and safety closed loop, ensures the continuity and safety of operations, reduces the risk of mis-disposal and mixed disposal, and improves operational efficiency and shift throughput.
Smart Images
Figure CN121553542B_ABST