Automatic control method and system for large-scale dispersed nodes based on unstable scene
By combining situational awareness and neural network models with mobility coefficients to classify people to be rescued, and constructing a cost matrix for optimal matching, the problem of uneven task allocation in large-scale decentralized rescue is solved, and low-cost and efficient rescue path optimization is achieved.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- THE NAVAL MEDICAL UNIV OF PLA
- Filing Date
- 2026-04-07
- Publication Date
- 2026-07-17
AI Technical Summary
In large-scale, decentralized rescue scenarios, existing technologies fail to effectively consider the different states of people awaiting rescue under unstable conditions, resulting in suboptimal allocation of rescue tasks and impacting rescue efficiency.
By acquiring the status information of people to be rescued through situational awareness, they are divided into three groups: the first group, the second group, and the third group. Different rescue matching is carried out based on these groups. The presence of floating objects is determined by using a neural network model. The movement coefficient is combined for detailed classification, and a cost matrix is constructed for optimal matching.
It enables optimized allocation of rescue missions in unstable scenarios, reduces overall rescue costs, and improves the scientific nature and accuracy of rescue efforts.
Smart Images

Figure CN122411547A_ABST