Corporate
Profile

Sales Pitch

Sensor IoT platform

Remote monitoring and control IoT system for power consumption and environmental data

When monitoring power consumption of equipment in facilities (facilities like hospitals and libraries, factories), or environmental data such as temperature, humidity, various gas concentrations, noise, vibration, and so on, all measurement data can be monitored remotely from a smartphone, tablet, or other information terminal via the cloud, without having to go near each piece of equipment. All measurement data can be remotely monitored in real time.

Moreover, data can be analyzed and displayed in easy-to-read formats such as graphs and tables, and alerts can be automatically sent to designated terminals when abnormal values occur. In addition, each sensor can be remotely controlled from the terminal, such as turning them on and off.

For IoT systems including cloud computing, the same system can be used for multiple facilities by changing sensors.

Furthermore, if this service is offered for a fee, the same system can be used to handle it, providing a highly efficient service.

We would like to work with IT companies that are considering new service delivery models using sensor IoT systems.

We are also interested in working with sensor manufacturers who are looking to expand their business by offering IoT solutions to their customers using their own sensors.

IoT system for predicting motor failures

Production lines have many pieces of production equipment such as motors. If they suddenly break down during production, it can cause extensive damage, such as a significant delay in the production plan due to equipment repair or replacement, disposal of work-in-progress, and so on.

Therefore, failure prediction is an urgent issue to avoid unforeseen situations.

We have developed a system that collects vibration, current, heat data, and so on from each production device and uses AI analysis to predict failures. In this system, an AI inference machine that can predict failures from acquired sensor data is mounted on an IoT gateway placed near the equipment. Decision data can be uploaded to the cloud with low power consumption through LPWA.

This allows the IoT gateway to operate for long periods of time (six months to a year) without battery replacement, even in locations without a power supply, and enables remote monitoring.

To improve predictive performance, additional training with new data at the server is required, and this large amount of data is uploaded via WiFi once a month.

The results of additional learning are again downloaded to the IoT gateway via WiFi, thereby reducing the amount of time spent on WiFi, which routinely consumes a large amount of power.

Product appearance inspection system

We developed an AI-based visual inspection system to eliminate the outflow of defects in the appearance of cast parts that had occurred when inspections were performed by humans.

We have realized a system that can detect appearance defects on multiple sides of a product and read its model number/lot number (characters difficult to distinguish due to shape distortion) with high accuracy and speed using a single industrial PC, based on image AI analysis.

The data from these products are uploaded and stored on a server or in the cloud via PLC. Appearance defects are linked to the model number/lot number to identify upstream processes that cause defects, and to contribute to yield improvement.

In addition, automation of visual inspection processes (lower cost), improved productivity through continuous operation, improved inspection accuracy that does not depend on human operators, stabilized accuracy, and easy line deployment both in Japan and abroad are expected.

We hope to work with various manufacturing companies that are seeking to reduce labor, lower costs, accelerate line deployment, and reduce appearance defects through visual inspection of their products.

This company is recommended by the following support organizations.

国立研究開発法人新エネルギー・産業技術総合開発機構（NEDO）

大阪信用金庫