AI Inference and PLC Data Mapping via Edge Device Intermediary
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing technologies face challenges in seamlessly integrating AI inference devices with automation controllers like PLCs, due to differences in device types, environments, and the need for complex data format conversions and communication protocol management.
Innovation Solution
A data interaction method that analyzes the topological structure of an AI model, matches output nodes and communication protocols, and generates source codes for both the AI inference device and the automation controller to facilitate standardized data interaction and format mapping.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If AI inference devices are integrated with automation controllers to enable AI functions, then the capability and versatility are improved, but the device complexity and difficulty of operation increase due to differences in device types, environments, and programming languages
Solution Approach 1:
The patent introduces an edge device as an intermediary between the automation controller and AI inference system. The edge device runs the AI model and handles data processing, while the automation controller maintains its existing functionality. This mediator approach allows AI capabilities to be added without modifying the controller's core system, thus resolving the contradiction between adding versatility and maintaining simplicity.
Solution Approach 2:
The system is segmented into distinct functional modules: the automation controller for control logic, the edge device for AI inference, and separate communication protocols for each layer. This segmentation allows each component to operate independently with its own optimized technology stack, reducing overall integration complexity while maintaining AI capabilities.
2Reliability
If data interaction is implemented through hard-coded protocols and custom programming, then communication reliability is improved, but the ease of operation and productivity decrease due to extensive programming expertise requirements
Solution Approach 1:
The patent implements a universal data interaction mechanism that supports multiple communication protocols (PROFINET, Modbus, EtherCAT) and programming languages (Python, C++, IEC 61131-3) through a unified framework. This universal approach maintains reliability by supporting established protocols while improving ease of operation by allowing engineers to work in their familiar languages without needing to learn new hard-coded protocols.
Solution Approach 2:
The system enables self-service through automated code generation and configuration tools that reduce manual programming requirements. The standardized interaction mechanisms allow the system to automatically handle data format conversion and protocol selection, reducing the need for extensive custom programming while maintaining communication reliability.
3Manufacturing precision
If AI models are trained and verified offline, then the manufacturing precision of the model is improved, but the loss of time increases due to format conversion requirements and inability to directly run on industrial devices
Solution Approach 1:
The patent implements preliminary action by pre-configuring standardized data interaction mechanisms and communication protocols during system design. Data format specifications and interaction frameworks are established in advance, allowing offline-trained models to be quickly deployed to edge devices without time-consuming format conversions or retraining, thus reducing deployment time while maintaining model accuracy.
4Productivity
If dedicated AI inference devices are used with hardware accelerators, then the productivity and processing speed are improved, but the device complexity and cost increase
Solution Approach 1:
The edge device serves as an intermediary that provides AI inference capabilities with appropriate hardware acceleration without requiring the automation controller itself to be complex. The edge device can be selected based on specific performance requirements, allowing productivity to be optimized independently of the controller's hardware complexity.
Data Source
AI summary
The teachings of the present disclosure include data interaction methods, apparatus, and systems for an Artificial Intelligence (AI) inference device and an automation controller. An example method includes: S1: analyzing a topological structure of an AI model, and acquiring multiple nodes of the AI model and interconnections of the multiple nodes; S2: matching an output node of the AI model based on a user input, determining a format of an inference result based on the AI model, and mapping the format of the inference result to the automation controller; and S3: matching a communication protocol for performing data interaction by the AI inference device and the automation controller based on the user input.