2results about How to "Timely conversion" patented technology

This invention discloses a coordinate system identification method, apparatus, electronic device, and storage medium. The coordinate system identification method includes: acquiring map messages and feature information contained in V2X information sent by a roadside device, and constructing a first road topology map based on the map messages; constructing a second road topology map based on a terminal map within a preset threshold range; determining the first position information of node information in the first road topology map located in the second road topology map based on the first road topology map, the second road topology map, and the feature information; determining the second position information corresponding to the node information in the first road topology map according to a preset map deflection rule; and determining the coordinate system to which the V2X information belongs based on the first and second position information. This invention enables rapid and accurate determination of the coordinate system to which V2X information belongs, facilitating timely conversion of V2X information to the coordinate system of the terminal, allowing the terminal to obtain accurate V2X information for application.

This invention discloses a multi-agent reinforcement learning-based supply chain network resilience optimization and resource allocation system. The system includes: a data preprocessing unit, an agent architecture construction unit, a pattern recognition unit, a non-cooperative game-theoretic allocation unit, a collaborative optimization allocation unit, and a dynamic switching output unit. The data preprocessing unit is configured to acquire the raw operational data and environmental state data of the supply chain network and perform preprocessing operations. The agent architecture construction unit is configured to build a multi-agent architecture for the supply chain. This invention relates to the field of supply chain network management technology. This multi-agent reinforcement learning-based supply chain network resilience optimization and resource allocation system uses a dynamic switching output unit to collect data in real time, update the agent and supply chain network states, and dynamically adjust resource allocation strategies based on a pattern switching mechanism to achieve adaptive switching between efficiency and resilience, adapting to complex and ever-changing supply chain environments.