Redefining Smart Factory with 5G

Next Gen CMI

Redefining Industry Solutions with 5G and Edge

 
September 7, 2020

 5G and Edge Computing: Piloting Smart Factory and Real-time Traffic Video Analytics

Since the past few years, enterprises are moving infrastructure to a centralized cloud, enabled by virtualization. This helps reduce time to market for new services and reduce total cost of ownership. Centralized cloud is apt for today’s use cases where ultra-low latency communication is not the predominant factor. However, in recent years, we are observing the emergence of new technologies such as augmented and virtual reality, autonomous factories, and IoT driven smart factories where data is increasingly being produced at the user end of the network. This in turn is increasing the demand for real time analytics and personalization at the edge itself. 

As public edge moves closer to the sites, it will allow enterprises to gain the benefits of better computing power in close proximity to end-users, without the need to own and operate associated infrastructure. 

5G and multi-access edge computing (MEC) will help enterprises by enabling solutions that exploit AI, IoT, analytics and cloud to deliver massive device connectivity, innovative technology experiences and near real-time automation.

TCS’ Communications, Media and Information Services (CMI) and Manufacturing business groups recently partnered with Verizon (5G MEC) and Amazon Web Services (AWS) Wavelength to test ultra-low latency applications for mobile devices on the newly launched AWS Wavelength. We tested two industry use cases – Intelligent real-time visual quality inspection for a smart factory and real-time digital (traffic) surveillance. Let’s take a closer look at these use cases to bring out the potential of exploiting edge and 5G. 

A connected digital ecosystem

# 1 Machine vision based defect detection in a smart factory:  A smart factory leverages AI, robotics, smart devices, smart computing, intelligence at the edge, and various self-healing systems. The use of cutting-edge technologies in communication and edge analytics makes the smart factories highly responsive to the demands of operational processes – a key ask for high-speed operations. For instance, videos from a smart factory assembly line, captured by IP based cameras, are transmitted using the Real Time Messaging Protocol (RTMP) protocols over the 5G bandwidth to servers in an edge zone. This reduces the time of transport to the processing server on edge with Graphics Processing Units (GPUs) deployed to process the videos. The video is then processed frame by frame and a processed frame is sent back to the console or device over 5G again, from which the machine operator who is monitoring the process, takes a decision. The decision module can also send a direct command from application to start or stop the assembly line or process. The response back to Programmable Logic Controllers (PLCs), or assembly line demands high latency which is supported by the 5G network. Leveraging 5G’s bandwidth and latency as well as proximity of plant to edge zones, can help move the workloads to edge zones efficiently. 

# 2 Real time digital  surveillance: Another fast-growing application in the physical security sector is CCTV cameras with AI capabilities. Amid the rising concern for public safety and security, governments around the world are pushing for better surveillance. 5G network provides a highly reliable, higher bandwidth, and low latency connectivity. 

The real-time video surveillance use case solution on edge enables intelligent processing and analytics of livestream video in ultra-low latency with a high degree of image and video resolution. The solution utilizes a deep neural network, deployed on edge cloud over 5G networks. It enables real-time detection of vehicle number plate and illegal parking to send notifications to owner or authorities. It offers the potential of scalable, robust and uninterrupted surveillance systems used for near real-time response in emergencies and public safety situations, as well as smart city applications. Instead of sending all video surveillance data to the cloud, multi-access edge computing reduces the security risk by processing data locally within the edge of AI enabled 5G network.

Moving intelligence to the edge for sharper, real time insights

5G network provides a highly reliable, higher bandwidth, and low latency connectivity for  - the smart factory and remote video real time digital surveillance solutions. The 5G-based surveillance solutions must be edge native and capable of fusing high-speed networks, driving more intense local processing, and is backed by an edge cloud infrastructure. The latter is capable of storing and analyzing data, streaming from a multitude of connected devices.

Edge cloud coupled with 5G’s high bandwidth, low latency, and increased connectivity ensures a firm step forward towards building truly connected smart factories with data flowing seamlessly between machines and enabling digital twin based industrial automation. Further, AI-enabled 5G cameras leveraging video analytics can be implemented for improving the traffic flow in real-time by detecting various traffic conditions.  The time is ripe to make AI powered real time decisioning at the edge a reality.  Find out more at the TCS Innovation Forum – A New Beginning: Reimagining Businesses in the 5G Era.

Sr. Consultant, Manufacturing Business Unit, TCS

Deepak is part of the CTO organization for TCS Manufacturing business group and leads its cloud/ AI vision offerings. A Mechanical engineer with 20 years of experience, he is also the product owner for CQA, a solution in the space of Eol quality inspection.

 

Chief Architect, Communications, Media & Information Services Business Unit, TCS

Sujatha, with over 20 years of cross industry experience, is a seasoned technology leader for digital business transformation programs.