HMI Feedback Loop for Industrial ML Model Training

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing industrial automation environments fail to effectively integrate user experience into machine learning models to effectively integrate machine learning models to effectively integrate user experience into machine learning environments to effectively integrate user experience into machine learning environments to effectively train user experience into machine learning environments to effectively integrate user experience into machine learning environments to effectively integrate user experience into machine learning technologies.

Innovation Solution

The integration of machine learning models into industrial automation environments is facilitated by connecting them with industrial controllers and HMIs, allowing user feedback to be received and utilized to enhance training through a feedback loop.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If machine learning models are integrated into industrial automation environments, then the ability to recognize patterns and automatically improve is enhanced, but the effectiveness of training these models with user experience data is insufficient

Engineering Contradiction:

Improvemachine learning model training capabilityVSAvoidtraining effectiveness

Core Design Contradiction:

Adaptability or versatilityVSReliability

Solution Approach 1:

The system implements a feedback mechanism where user interactions with the HMI are captured and fed back to retrain the machine learning model. The model generates outputs displayed on the HMI, users provide feedback on these outputs, and this feedback is used to continuously retrain and improve the model, creating a closed-loop system that enhances training effectiveness.

Inventive Principle:

Principle #23Feedback

Solution Approach 2:

The machine learning model performs self-improvement by automatically retraining itself based on user feedback collected through the HMI interface. The system enables the model to service its own training needs without external intervention, continuously adapting to user preferences and improving its outputs autonomously.

Inventive Principle:

Principle #25Self-service

2Ease of operation

If control systems are used to drive industrial operations, then operational control is achieved, but integration of user experience into machine learning training is ineffective

Engineering Contradiction:

Improveuser interaction capabilityVSAvoiduser experience data utilization

Core Design Contradiction:

Ease of operationVSLoss of information

Solution Approach 1:

The HMI captures user feedback on machine learning model outputs and feeds this information back to the training system. This feedback loop ensures that valuable user experience data is not lost but instead utilized to improve future model predictions and outputs.

Inventive Principle:

Principle #23Feedback

Solution Approach 2:

The HMI serves as an intermediary between the machine learning model and the user, facilitating the collection and transmission of user feedback. It mediates the interaction by presenting model outputs to users and capturing their responses, which are then used to improve the model.

Inventive Principle:

Principle #24Intermediary (Mediator)

3Productivity

If machine learning models process data at fast speeds, then processing efficiency is improved, but the integration of user feedback into model training is insufficient

Engineering Contradiction:

Improvedata processing speedVSAvoiduser feedback integration

Core Design Contradiction:

ProductivityVSAdaptability or versatility

Solution Approach 1:

The system maintains continuous operation by processing data at high speeds while simultaneously and continuously integrating user feedback into model training. The feedback loop operates continuously, allowing the model to adapt in real-time without interrupting the fast data processing workflow.

Inventive Principle:

Principle #20Continuity of useful action

Solution Approach 2:

The system prepares for feedback integration by having the training infrastructure ready to process user feedback as it arrives. The model is pre-configured to accept and process feedback data, enabling seamless integration without disrupting the fast processing speed.

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS12518211B2Obtaining user feedback on models via a human machine interface (HMI)

Publication Date: 2026.01.06 ROCKWELL AUTOMATION TECH INC

AI summary

Various embodiments of the present technology generally relate to industrial automation environments. More specifically, embodiments include systems and methods to train machine learning systems based on user operations in an industrial automation environment. In some examples, a Human Machine Interface (HMI) component displays a machine learning output indicating a training state of a machine learning model on a user interface and user feedback regarding the training state. A machine learning interface component weights feature vectors based on the user feedback and supplies the weighted feature vectors to the machine learning model. The machine learning interface component receives another machine learning output that indicates an updated training state for the model. The HMI component displays the output indicating the updated training state of the model, receives additional user feedback regarding the updating training state, and transfers the additional user feedback to the machine learning interface component.