Method, apparatus, device, and medium for ai inference in industrial automation system

Centralizing AI models in a centralized inference server addresses maintainability and resource waste issues by enabling efficient deployment and scalable hardware use in industrial automation systems.

WO2026112784A1PCT designated stage Publication Date: 2026-06-04SIEMENS AG +1
View PDF 5 Cites 0 Cited by

Patent Information

Authority / Receiving Office
WO · WO
Patent Type
Applications
Current Assignee / Owner
SIEMENS AG
Filing Date
2024-11-26
Publication Date
2026-06-04

AI Technical Summary

Technical Problem

Existing industrial automation systems face issues of low maintainability, waste of hardware resources, and poor scalability due to the integration of AI models in edge devices, which require manual maintenance and upgrades on production lines and consume significant hardware resources.

Method used

Centralizing AI models in a centralized AI inference server that receives sensor data from multiple workstations, executes inference, and sends results back, allowing for efficient deployment, maintenance, and scalability of hardware resources by sharing acceleration units across multiple models.

Benefits of technology

This approach reduces maintenance efforts, optimizes hardware resource usage, and enhances scalability by enabling the addition of more acceleration units as needed, improving the efficiency and flexibility of AI model deployment in industrial automation systems.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN2024134578_04062026_PF_FP_ABST
    Figure CN2024134578_04062026_PF_FP_ABST
Patent Text Reader

Abstract

Embodiments of the present disclosure disclose a method for AI inference in industrial automation system, comprising: deploying at least one AI model in a centralized AI inference server; receiving, at the centralized AI inference server, sensor data from a plurality of workstations in production lines of the industrial automation system; executing, by AI model in the centralized AI inference server, inference using the received sensor data to generate inference results; sending the inference results to the respective workstations.
Need to check novelty before this filing date? Find Prior Art