A multi-robot cooperative handling method and system
By dynamically identifying and enhancing the observation capabilities of key robot nodes, optimizing the selection of team representatives and information aggregation, the problems of task mismatch and low information interaction efficiency caused by individual differences of robots in automated warehousing are solved, thereby improving transportation efficiency and system robustness.
Patent Information
- Application Number
- CN202610842043.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-06-11
- Publication Date
- 2026-07-17
AI Technical Summary
In automated warehousing scenarios involving parallel receiving and dispatching of goods at multiple locations and dynamic updates of handling tasks, differences in the load capacity and motion performance of individual robots lead to task mismatch and low efficiency in information exchange. Existing methods fail to dynamically identify key nodes and provide targeted perception, resulting in uneven allocation of transport capacity.
By acquiring the robot's local observation embeddings and relationship graphs, information entropy, structural centrality, attention diversity, and intrinsic feature strength are calculated to dynamically select key robot nodes, enhance local observation embeddings, and optimize team representative selection and information aggregation by combining a hybrid strategy of probabilistic sampling and deterministic selection.
It improved information aggregation efficiency and decision-making accuracy, enhanced adaptability and robustness in complex environments, optimized the selection of team representatives, and improved the quality of cross-team communication and transportation efficiency.
Smart Images
Figure CN122414727A_ABST