By Liran Bar, VP Business Development, Hailo
As evolving industrial demands and technological breakthroughs bring new advances in Industry 4.0, the global smart factory market is set to reach $244.8 billion by 2024, compared with $153.7 billion in 2019, according to research by Markets and Markets.
Major drivers of this growth will be innovations in industrial robotics, IoT, and smart automation. But many computing solutions powering these advances still lack sufficient processing power to meet target performance levels. Current solutions simply cannot keep up with round-the-clock, multi-device automation and connectivity.
Indeed, more efficient processing at the edge is vital in order to usher in the next generation of Industry 4.0. New, innovative, AI-powered technologies are the very thing the Industry 4.0 revolution requires to reach peak performance.
The Need for Hyper-Efficient Processing
In modern factories, where time directly translates into money and where every split-second counts, it is essential that multiple video streams from the production floor are rapidly acquired and processed in real time. Processing at the edge rather than in the centralized cloud translates into significant cost reductions along with swifter, more efficient processing for tasks such as inspection, quality assurance and better safety measures for human interaction with machines.
The outdated computer architecture manufacturers have traditionally relied on to power AI applications has resulted in low processing throughput, high latency, high power consumption, and high costs. Today’s smart factories need solutions that don’t compromise on either performance or cost. The same goes for smart retail and cashier-less stores like the Amazon Go using ceiling-mounted cameras for visual intelligence and monitoring inventory.
From a cost-competitiveness perspective, the advantage of edge processing is clear. A solution that processes and analyzes single or multiple streaming camera input feeds in real time, all at the edge, means lower costs and higher efficiency.
The Next Generation of Industrial IoT
This is precisely what Hailo, along with Foxconn and Socionext, are enabling through the development of a new generation AI processing solution for video analytics at the edge. Bringing to bear each of our unique sets of expertise – Foxconn, a global smart manufacturing powerhouse, Socionext, a developer of leading video and imaging systems solutions and Hailo, a pioneering AI chipmaker – the three companies have joined forces to build a robust, high-efficiency product capable of processing and analyzing over 20 streaming camera input feeds in real-time, all at the edge.
The result is a high-density, low-power, complete local VMS server, ensuring top performance for video analytics and privacy, including image classification, detection, pose estimation, and various other AI-powered applications – all in real time.
The solution fuses Foxconn’s high-density, fan-less, and highly efficient edge computing solution BOXiedge™ with Socionext’s high-efficiency parallel processor SynQuacer™ SC2A11 and the Hailo-8™ deep learning processor. BOXiedge™’s fan-lessness is a crucial advantage, as this prevents the accumulation of dust that tends to rapidly degrade processors equipped with fans, leading to poor performance and high replacement costs. Combine this with the Hailo-8™’s ability to offload deep learning-based applications, and the result is a peerless, next-gen AI processing solution for Industry 4.0.
Foxconn has already deployed several in-house AI solutions on different production lines, leading to an improvement in reporting accuracy from 95% to 99% and a reduction of at least one-third of operating costs for appearance defect inspection projects. The solution ensures better factory safety as well – if a human worker approaches a potentially dangerous machine, for instance, the machine will stop working automatically.
This unique partnership between Foxconn, Hailo and Socionext has created the next generation of edge processing for Industry 4.0. Enabling processing at the edge, rather than in the cloud, translates into significant cost reduction along with more efficient processing – all crucial to improving the bottom line.