From prototyping to commercial level end to end IoT edge solutions


Improving Efficiency, preventing risks and delivering result with connected IoT devices.

Industry 4.0 is an era in which information technology is used to promote industrial change. In such an intelligent era, technology can improve our production efficiency in multiple dimensions. Traditional industrial field data collection usually uses sensors to deploy independently, and dedicated staff is responsible for reading the statistical data point by point.

For resource optimization and industrial efficiency improvement, many industrial production systems are gradually carrying out improvement plans.

We have learned from some manufacturers who have worked closely with M5 that before contacting M5 products, the manufacturer's internal R&D engineers used the water-proof protective box + Arduino PCB board scheme. Although it can fit basic functions demand, the bare PCB has always made people unsafe. And every new feature addition and modification is a challenge to the time cost. In addition to the ugliness, the way to add a protective box is also a cost control test.

· Lack of efficient data collection methods
The original way of reading the meter through a person, for each node, record and import the table.
· Node deployment cycle length, difficulty
The installation and deployment of equipment require frequent debugging and lacks complete software and hardware service facilities.
· Lack of equipment monitoring
Unable to know the equipment failure and alarm information at the first time, causing equipment maintenance delay. Lack of intuitive monitoring and alarm push channels for equipment.
· Poor equipment operation reliability
Without a complete protection structure design, the PCB circuit is easily disturbed by the outside world, and there are hidden safety hazards.

Based on the above common problems, after the actual evaluation and use experience of various market products, many R&D engineers chose M5, a brand with a complete software and hardware system.

Compared with the traditional recording method, the addition of M5Stack hardware products has accelerated the transformation of Industry 4.0 in the production environment data collection method.

Simple, reliable, and highly integrated master control equipment and sensors can quickly and efficiently deploy on-site nodes and reliably collect industry on-site data.

The minute the pressure of each valve and the number of movements of the operating bed are accurately measured by sensors and embedded master control devices and gathered through the cloud to the visualized background panel, there is no need to check one by one on the person.

Easily control the working status of all equipment and the data change curve.

· Data Acquisition in Cloud
Field device data is submitted to the cloud for statistics to form a data change curve. There is no need to have a fixed site record, which is more effective.
· High-efficiency node deployment
Provide a convenient graphical programming platform on the software. From development to deployment of new features, it only takes a few days!
· Real-time sensor monitoring
Various monitoring methods, such as optical, pressure, and electric, can quickly locate faulty equipment location and fault type through remote push warning.
· Stable sensor operation
The structure and circuit design tailored specifically for the industry scenes, reliable equipment continues to operate.


Using sensors to collect data and using edge computing devices to analyze data, helps automate farming and breeding sectors.

Livestock has the longest history of traditional industry.

The technological development of Industry 4.0, to a certain extent, has also promoted the industrial transformation of the animal husbandry industry.

How to make livestock management more “intelligent” “cutting-edge livestock” “smart farm”. This kind of vocabulary appeared in the eyes of people.

In order to improve the efficiency of work, many farmers have also begun to develop IoT solutions that can be applied to their own animal husbandry operation system.

Combining the current technology to optimize the traditional breeding scheme, it has a considerable promotion effect on the management of the livestock life cycle and sales.

Our project manager learned from a pig breeding base in Japan that the basic Th pig breeding requires a macroscopic process of raising, achieving weight detection, screening, and selling.

In view of the optimization of the efficiency of animal husbandry nuclear operations, combined with actual operation scenarios.

· Pigs are sorted according to their status
Rough judgment is easy to make mistakes and missed checks.
· Pig weight compliance detection
The above-mentioned difficulty is relatively high, and the work efficiency is low.
· Human monitoring of feeding status
Involvement of inspections, for large-scale farms, opening up is a difficult problem
· Livestock statistics
The statistics of the number of livestock is very important, but in the high-frequency statistics, it is cumbersome and consuming.

Through the use of pain point analysis, M5 puts forward some suitable solutions for some of the current problems.

For the preliminary sorting of the pigs that meet the standards, we used the ESP32-CAM wireless WIFI camera program to take pictures of the livestock in the breeding pen at regular intervals and upload them to the cloud to perform AI analysis on the individuals who meet the standards. Tracking marks, the automatic output of corresponding list sets, prompting the breeder to weigh the individuals in the set for the next step of testing.

Deploy the embedded "weighing area" at a fixed point in the breeding circle (built-in weighing sensor and high-precision ADC conversion core, combined with the main control CORE to achieve wireless data upload), only need to pull the livestock to Specify the area to automatically record the weight information, form a stream-type operation, and speed up the work efficiency.

· AI sorting to be tested
Wireless WIFI camera + cloud AI analysis plan, a form, feedback all results.
· Deploy a dedicated area for weighing
By delimiting the weighing area, the livestock is driven into the inspection area to form a weighing queue.
· Real-time monitoring of feeding status
Through the camera image, the status of the feeding circle is detected in the background, and the number of inspection personnel is reduced.
· Intelligent statistics of livestock
Use camera images to filter through algorithms for efficient statistics.


Obtain and analyze data about customer behavior; Automate shopping process and optimize shopping experience.

Out-of-stock (OOS), also known as Stockouts, are among the most frustrating experiences for online and in-store shoppers.

Retailers are missing out on nearly $1 trillion in sales because they don't have on hand what customers want to buy in their stores, according to a study conducted by IHL Group.Shoppers encounter out-of-stocks in as often as one in three shopping trips, according to the report.(source)

And almost a third of shoppers ended up turning to Amazon when the product they wanted wasn't in stock at their local store.

· Loss of retail orders during an OOS period
During the OOS period, it is easy to lose potential retail orders.
· Lack of data analysis source
Record and analyze the impact of shelf area on merchandise sales.
· Labor resources consuming
A large number of shelf management staff need to repeat inspections and replenishment of OOS products.

Applying with Low-power consumption Timer Camera launched by M5Stack, it can effectively take pictures of shelves at regular intervals, connect to the WIFI network, upload to cloud server, and analyze the inventory of shelves with deep learning algorithm. Instantly analyze and alert the staff with the OOS products and locations.

Compared to the traditional operation procedure, M5Stack hardware solution simplifies the inspection process for store managers, and avoids the potential loss of orders caused by products shortage, from store to store. That is also a big help on freeing up labor resources.

Simple, reliable, and highly integrated TimerCAM can instantly and efficiently deploy on-site to monitor on-shelf data.

· Smart on-shelf inventory replenishment alert
The system is applied for timing identification, alerting the staff for OOS products and shelf locations, to replenish products in time.
· Data Statistics/Graph Analytics
Record and analyze the impact of shelf location to product sales.
· Labor resources are allocated reasonably
Free up more working hours for labor resource.
· Low-power consumption Wifi camera on-shelf monitoring
The ultra-low power consumption camera is programmed to wake up regularly, and the sytem can work continuously for more than a year without replacing battery.