A Closer Look at Wistron’s ZigNeurons
Cars and light rail breathe life into the city, but they also create problems in areas such as environmental pollution and safety of drivers, passengers and pedestrians. Recently however, with the advancement of technology and connectivity, these problems can now be solved. Wistron Intelligent Mobility Solutions (IMS) integrates AI artificial intelligence and IoT smart networking technologies to provide safer, more efficient, and environmentally friendly mobility solutions.
Current AI Landscape in the Transportation Industry
Within the transportation industry, AI artificial intelligence is already used in self-driving cars, by law enforcement, for vehicle tracking and fleet operation optimization. It is primarily utilized to perform highly repetitive work, or inspection in hazardous work environments, as well as to perform complex computations. However, the technical implementation of AI still faces many challenges when it is actually introduced.
Traditional AI is difficult to implement within the fast changing transportation industry. When encountering unexpected situations or changes in the surrounding environment, problems such as detection errors start to occur; if AI is to be applied in different application scenarios, it must spend a lot of manpower and resources repeatedly to train and fine tune the AI. Faced with such difficulties, the Wistron IMS team has used its strong industry experience and technical prowess to establish the “ZigNeurons” platform.
Upgrading Traditional AI Applications to a Continuously Evolving AI – Rail Inspection Example
ZigNeurons provides a platform for AI application model development and model evolution. Through Adaptive Machine Learning, the AI model can continuously collect data based on changes in objects and the environment, continuously train and update the AI model, and then continuously revise it. Through this power to continuously evolve, the AI model can provide more flexible and diverse solutions.
According to the statistics of the rail transportation, there were an expected 3,374 train delays of more than 5 minutes in 2021, affecting 14,531 trains, of which 2,674 delays (accounting for 79.25%) were mainly due to “equipment failure”. If equipment failures can be effectively eliminated, it means that the situation of train delays can be effectively reduced, and the riding experience of passengers can be improved at the same time.
In the past, equipment maintenance was mostly carried out manually. This repetitive and tedious maintenance work is time-consuming and prone to errors. Now, with the assistance of the ZigNeurons platform and its ability to learn and improve over time, it can effectively avoid machine damage caused by human negligence during inspection and greatly reduce train delays and cancellations.
Why is the ZigNeurons platform able to do things that traditional AI artificial intelligence cannot achieve?
Four ZigNeurons qualities are worth highlighting:
- Efficient and easy-to-use AI platform services
The ZigNeurons platform has two key characteristics, it is “efficient” and “easy to use”. It adopts the development method of continuous learning and training to promote the continuous optimization and automatic update of AI models. The interface offers a visual presentation of the smart model system that allows users to get started quickly. There are also comparisons of iterative development results between different training result versions which make the training results clear at a glance.
- Smart rule settings alleviate worry about system integration
The ZigNeurons platform has a variety of alert settings. Users can set various smart rules for different AI applications and monitoring devices. For example, sensors can be set to perform different inspection items during peak or off-peak hours, and they can also be integrated with existing enterprise resource planning systems and directly transmit the notifications to the duty personnel. The end result is that users no longer have to worry about the old and new systems, or to learn the logic of the new system.
- Monitoring of edge computing power ensures reliable system uptime
Image recognition AI applications, especially in environments where small objects move quickly, need to use high-definition video for recognition, which consumes a huge amount of edge computing power. Therefore, the status of edge computing power is a major factor that affects the detection speed and system operation. The key point is that when the edge computing power is continuously consumed to a point where it cannot meet the demands of AI applications that are running simultaneously. The speed of detection will be degraded and it may cause delays or even miss triggering an alarm completely. The computing power management and monitoring mechanism of ZigNeurons automatically calculates the optimal use of computing power based on any newly added AI applications, and adjusts the edge computing power to ensure the detection function can continue to work.
- Flexible expansion mechanism meets the various needs of enterprises
ZigNeurons adopts the scalable Kubernetes architecture so the system can quickly adjust according to the needs of an enterprise. For example, during operating hours, it uses crowd detection to monitor; during non-operating hours, it can route computing power to object intrusion detection. It can be set up to make adjustments at any time. It can even assign different devices to deploy different AI applications, reduce unnecessary detection computing power when necessary, react to various emergencies, and deliver an overall better system functionality.
Additional applications of Wistron IMS
The Wistron IMS team has solutions that go beyond assisting in the inspection of vehicles. The scope extends to fleet operation and management, which can assist logistics and transportation companies to strengthen the training of novice drivers, improve fleet operation efficiency, reduce costs, achieve energy conservation and help them meet their carbon reduction goals.
Please contact us if you want to learn more about our smart mobility solutions.
