The new company, Advanced Vision Solutions Co., Ltd., plans to begin operations in October 2026, pending regulatory approvals.
Its focus will be on developing solutions capable of capturing and analyzing visual information directly from image sensors using edge AI technologies.
The goal of this architecture is to gather information on production conditions, product quality, and machinery status, and use that data to automate decision-making and control processes.
AI Directly in Vision Sensors:
Vision sensors use cameras to capture images of products, components, and other elements on a production line. From these images, they can determine characteristics such as the state, position, or specific properties of the objects.
By connecting with industrial automation systems, the information obtained can be used to control machinery or activate different processes without the need for constant human supervision.
The integration of Edge AI allows some of the analysis to be performed near where the data is generated, instead of relying exclusively on processing on external servers or centralized platforms.
This approach can reduce the volume of information that needs to be transmitted and decrease the time required to interpret the data and generate a system response.
Industrial Control Based on Visual Information:
The planned solutions will combine image sensor technologies and Edge AI with control systems used in industrial automation.
Proposed applications include monitoring production conditions, quality inspection, analyzing the operational status of equipment, and gathering information for maintenance purposes.
The system will be able to extract relevant information from images and connect it to the systems responsible for making decisions or executing actions on the machinery.
In this way, visual data will no longer be used solely for inspection tasks but will become an integral part of the control processes of industrial facilities.
Multimodal Production Systems:
The project also envisions combining information from vision sensors with other types of data generated in manufacturing plants.
This integration will enable the development of multimodal systems that simultaneously analyze different sources of information to detect process changes, anomalies, or early warning signs of potential problems.
Combining multiple sources can be particularly useful when a situation cannot be reliably identified using a single type of sensor.
The data obtained could be used to adapt certain machine and process parameters in real time, as well as to develop predictive maintenance systems.
Physical AI Comes to Industrial Automation.
Another technology linked to the project is Physical AI, a concept that refers to artificial intelligence systems capable of interpreting information from the physical environment and using it to make decisions that are subsequently translated into actions.
In a factory, this approach can combine cameras and other sensors with industrial control systems to observe what is happening on a production line, interpret the data, and modify the operation of the equipment.
The goal is to move toward processes capable of reacting autonomously to changes in manufacturing conditions, with less need for human intervention.
These technologies can be applied to tasks such as automatic inspection, quality control, anomaly detection, equipment monitoring, and predictive maintenance.
The creation of Advanced Vision Solutions thus reflects the growing convergence between machine vision, edge AI processing, and traditional industrial automation systems—three technologies that are playing an increasingly important role in the development of factories with more autonomous processes.
